Tracing Business Impact Beyond The Click

A field guide for SEOs, AEOs, and Content Marketers: Move beyond clicks to investigate organizational impact, note where else consequences land, and work with stakeholders collaboratively to establish what the evidence supports for your work growing value in the zero click era.

THE OPERATING SEQUENCE

Bring a business question, a reported persistent problem, a result from another guide, or simply a REASON to start an inquiry into business impact. This method helps you investigate what happened, what information might have contributed, and what the evidence lets you say or do next.

Test a change only when warranted (E — CHANGE + LEARN). If an information change follows from the investigation, define what should improve and what must not get worse, observe what happens, then return to step 4 to interpret the result.

Find a thread, if you need one (A — SCAN). Look for a specific instance of work, burden, opportunity, or change worth understanding. If you already have one, begin at step 2.

Define the inquiry (B — SPEC). State what was observed, whose work or result matters, which decision this investigation must inform, and what remains uncertain. Do not assume the cause is informational.

Investigate the next consequential uncertainty (C — FOLLOW). Go to the people and records closest to it. Inspect what they show, how the records were made, where access or definitions limit them, and what further evidence would change the decision.

Place the result (D — LOCATE). State what was established, the evidence and scope behind it, what remains unknown, and the strongest claim it supports. Decide whether to use the result, investigate another question, collaborate with another owner, leave it bounded, or reframe the inquiry.

WAYS IN

Already have a reported problem, observed result, or evidence carried from AI Visibility or Information Design? Begin at B with that result and its current limits. Looking for a consequential question to investigate? Begin at A. Interested specifically in customer relationship value or cost? Use Appendix A — Optional focused scan: Customer relationship value and cost. Working with an assistant? Start with the concern and records you have; see “Using This Guide With an Assistant.”

HOW THE WORK MOVES

C and D may repeat: an answer may reveal a new uncertainty worth investigating. A is unnecessary when you already have a thread. E is optional; a useful investigation may end in a bounded conclusion with no information change.

GUIDE MAP

The Guide Map answers “Where do I go to do this?” The Contents pages that follow show what is in the book and where.

ORIENTATION

WHY THIS GUIDE EXISTS

Search and Content practitioners have traditionally relied on visibility, traffic, and conversion as convenient observation points. Those signals still matter, but they do not define where useful work or business consequence begins and ends. Information can affect fit, implementation, support demand, expert attention, rework, risk, retention, capacity, customer economics, and other parts of the organization that may never show up cleanly in a click path.

The click was an observation point, not proof of value. When it becomes less reliable, the value it stood near does not disappear; it becomes harder to see from where you were standing. This guide is about choosing new places to stand.

It helps you investigate organizational consequence without turning every business problem into an information problem and without stretching a close-in result into a stronger downstream claim.

You may need to investigate work outside your role or records you do not control. Begin with what you observed, why it may matter and in what decision, for which person or persons, and what you do not yet understand. Ask the people closest to the work about its constraints, record definitions, prior decisions, and other possible explanations. Keep their response and any roadblock visible as you continue the inquiry. A repeated difficulty or measured change may justify investigation; it does not by itself establish its cause, cost, or business impact.

WHAT SEARCH PRACTICE ALREADY GAVE YOU

If you were formed by Search, Content, or AI-visibility work, your home position is the interface between information-seeking behavior and the information environment. You have spent years watching what people and systems look for, what they find, what they fail to find, and what happens when the environment changes.

Much of that equipment travels.

Two things do not travel automatically. A Search signal is not automatically a human need or business problem. And your authority does not expand merely because the investigation moves into another function. The people who understand the work, the people who own the data, and the specialists whose judgment the work requires remain essential sources.

Your equipment can travel. Your claims should travel only as far as the evidence.

WHERE TO LOOK FOR A BUSINESS THREAD

Information can matter through a customer’s decisions and outcomes, and through the work an organization does to serve them. Start with a task, burden, risk, or result that people can describe: a difficult review, repeated clarification, rework at a handoff, a return, a support escalation, necessary specialist scrutiny, or a result noticed after a change. Ask who experiences the work, what is actually happening, which person or record can help verify it, and what the work produces, costs, or protects. Then investigate whether information is plausibly involved alongside other explanations. These are places to look, not impacts to claim.

Possible threads include a security review that needs evidence about an industrial gateway’s support lifecycle; a proposed pharmaceutical service engagement whose scope must survive procurement and transition; a product return where compatibility information may have mattered; or an internal quality review requiring evidence from several teams. Each is an investigation starting point, not a demonstrated business result. If the question specifically concerns customer value and cost across a relationship, use the Customer Relationship Value & Cost Map in Appendix A as a focused scan. Otherwise use A — SCAN to find a thread, or bring an existing thread to B — SPEC.

USE THIS GUIDE WHEN

EVIDENCE BOUNDARIES

An entry point is not a conclusion.

A query can show that people searched. An AI answer can show what the system represented. A support ticket can show that a problem occurred. A manager can report that work feels quieter. A business measure can show that the measure moved. Each gives you somewhere to begin. None, by itself, establishes the full human need, the cause, or the organizational consequence.

Use an evidence state when the strength of a finding, requirement, interpretation, expected effect, or handoff matters to a decision. You do not need to classify every sentence.

Provenance (the source history of your data) remains separate. Record where the evidence came from.

A human report is legitimate evidence. It does not by itself establish prevalence, size, durability, or cause. A business measure records its construct, not automatic improvement. A model is not an observation.

THE CITATION LABS FIELD-GUIDE FAMILY

Start anywhere. The next unresolved question determines the guide.

AI VISIBILITY: DIAGNOSE, INTERVENE, RE-TEST investigates the AI answer environment: what systems find, retrieve, cite, represent, omit, and recommend, and whether an AI-facing intervention changed those conditions.

INFORMATION DESIGN FOR HUMANS AT WORK investigates human work and the information environment: what people are trying to accomplish, what information must enable, where the environment falls short, what should change, and whether the change helped.

TRACING BUSINESS IMPACT BEYOND THE CLICK investigates whether a bounded information or human-work result produces a meaningful organizational consequence – or whether an observed business burden or opportunity has an information component worth investigating.

QUESTIONS, NOT ARTIFACTS  The guides own questions and evidence, not pages, resources, or systems. The same artifact may appear in more than one guide, but each kind of effect must be established separately.

HANDOFF RULE  When work moves, carry the established result at the strength already earned, together with its evidence state, evidence and provenance, material conditions, important limits, and the next uncertainty or hypothesis. Do not upgrade it on arrival.

ANOTHER DOMAIN  Product truth, Engineering judgment, Finance definitions, Legal or regulatory interpretation, formal causal estimation, and other specialist questions belong with the person or method closest to them.

THE WORKING RECORD

USING THIS GUIDE WITH AN ASSISTANT

Begin with your assisted entity’s concern, result, or decision in their own words. State what is reported or observed and what remains uncertain. If there is no specific thread, help find one through A — SCAN; use Appendix A only when customer-relationship economics is the useful lens. If a problem or result is already known, begin at B — SPEC and preserve any carried-forward result at its existing strength.

Ask for the next person, case, record, or comparison that could change the decision; check record integrity and plausible other explanations before making a claim. Ask only for information needed for the next consequential step. Do not imply that you inspected a record, interviewed an owner, or verified a fact that you could not access.

Explain in ordinary language what the evidence supports, what remains unknown, and what to check or decide next. Identify who owns a decision or handoff, and keep their response or blocker visible. Use formal Readout fields when a durable record, review, or handoff is useful, and explain any method term you use.

An assistant can generate hypotheses and organize the investigation, but it cannot supply facts about the organization that people, records, or the work itself have not established.

For each exchange, identify the decision the person is trying to make and the strongest result already established. Choose the next uncertainty whose answer could change that decision. Ask for the nearest person, case, record, or comparison that could answer it, checking access and record quality where relevant. When evidence arrives, state what it establishes, what remains open, and whether to investigate further, use a bounded result, collaborate with another owner, or test an information change. If evidence is unavailable or unreliable, state what cannot yet be concluded and suggest a feasible next check. Keep the working record current and explain the conclusion and next action in ordinary language.

You do not need to choose a section of this guide before asking for help. Start with the business question you have and the evidence you can share. The assistant should find the right route, state what is known and still uncertain, and ask for the next useful person, record, or comparison without inventing facts about your organization.

You can start with:

“We changed [a page, information path, or AI-facing resource] and observed [result]. Can we say it affected [business result someone cares about]? What evidence would we need?”

“Sales says prospects keep asking our engineers the same setup questions before they buy. Could our site be part of the problem?”

In ordinary conversation, explain what the available evidence shows, what remains uncertain, and what would help decide the next step in the practitioner’s own terms. Apply the guide’s evidence and claim discipline throughout. Use the formal Readout fields when the person requests a structured record or a handoff, and explain any method label that appears in a response.

A — SCAN

WHERE MIGHT INFORMATION ALREADY MATTER AROUND HERE?

Use A when you need help finding a worthwhile thread. Scan for places where information may affect consequential work or organizational burden.

Common places include fit and selection; evaluation and persuasion; Sales and expert attention; onboarding and implementation; use and success; Support and recovery; rework and coordination; risk, quality, and warranted scrutiny; retention, expansion, and exit; organizational knowledge; and Search, AI, and public information.

You can scan broadly or start somewhere you already have access. A colleague you trust in Support, Operations, Product, Implementation, Finance, Compliance, Sales Engineering, Engineering, or Customer Success may be a better doorway than the theoretically best opportunity. A is a conversation tool as much as a worksheet.

If customer relationship value and cost are the terrain you want to understand, use Appendix A — Optional focused scan: Customer relationship value and cost.

For anything interesting, ask:

REAL? Am I pointing at something somebody actually sees or experiences?

CONSEQUENTIAL? Could it matter to work, success, burden, risk, capacity, or value?

CAN I GET CLOSE TO IT? Can I reach people, cases, records, or systems that would help me understand it?

INFORMATION-PLAUSIBLE? Could information credibly be involved, even though I do not yet know whether it is?

Pick one that is real enough to investigate, consequential enough to care about, and close enough that you can learn something. These are places to look, not outcomes to claim.

Take it to B.

B — SPEC

TURN ONE THREAD INTO A REAL INVESTIGATION

Your first task is not to explain the problem. It is to become precise about what is happening, what consequential work is involved, and what remains uncertain.

B1. WHAT BROUGHT YOU HERE?

Name the doorway: a Search, AI, or analytics signal; a person or team you already know; a business problem, burden, opportunity, or result; a customer situation; evidence carried forward from AI Visibility or Information Design; or something else.

If you arrived with carried-forward evidence, do not upgrade it. Its established result enters at the evidence state already earned. The hypothesis that came with it is what you investigate, not what you inherit as fact.

B2. WHAT ARE WE ACTUALLY SEEING?

Describe the event, condition, report, or result without explaining why it happens. Record the basis/source and evidence state.

A useful check: could another reasonable person inspect the same evidence and agree that this occurred, even if they disagreed completely about why? If not yet, get closer.

B3. WHAT CONSEQUENTIAL WORK IS HAPPENING HERE?

Name the person or group, what they are trying to accomplish, the real conditions or constraints around the work, and what good looks like.

Do not reduce consequential work to a decision. People may be trying to understand, compare, verify, implement, coordinate, troubleshoot, recover, comply, persuade, escalate appropriately, or complete another consequential task.

Name why it could be consequential: customer success, fit, Sales work, expert attention, Support, implementation, rework, quality, risk, retention, capacity, cost-to-serve, or another result. You are naming a possible consequence, not claiming that the starting signal caused it.

CALL INFORMATION DESIGN WHEN NEEDED  Name the consequential work far enough to spec this inquiry. If establishing what information must enable or diagnosing the human-facing information gap becomes the investigation, move to Information Design for Humans at Work.

B4. IS INFORMATION ACTUALLY AN IMPORTANT PART OF THE CONDITION?

Ask:

IF EVERYONE INVOLVED HAD THE RIGHT INFORMATION, AVAILABLE AT THE RIGHT TIME AND IN A FORM THEY COULD USE, WHAT IMPORTANT DIFFICULTY WOULD STILL REMAIN?

Keep both sides visible.

WHAT COULD BETTER INFORMATION PLAUSIBLY CHANGE?

WHAT WOULD STILL REMAIN BECAUSE OF PRODUCT CAPABILITY, RELIABILITY, ENTITLEMENT, POLICY, WORKFLOW, SKILL, STAFFING, CAPACITY, NECESSARY EXPERT JUDGMENT, OR ANOTHER CONDITION?

Record your current read:

INFORMATION APPEARS IMPORTANT.

MIXED.

INFORMATION IS PROBABLY NOT THE MAIN CONSTRAINT.

UNKNOWN — GET CLOSER BEFORE DIAGNOSING.

A legitimate result is: INFORMATION IS NOT THE IMPORTANT CONSTRAINT HERE.

B5. WHO OR WHAT IS CLOSEST TO THE REALITY?

A person? A customer? An actual case? Support records? Calls? Notes? Workflow records? Product or operational data? Search or query patterns? AI answers? Complaints? Reviews? A specialist?

Name the lightest credible way you can get closer.

Before opening tickets, CRM records, call notes, task histories, or other individual or client records, confirm that you have appropriate access and permission. Prefer patterns to individuals and minimize or anonymize individual-level detail when it is not necessary to the question.

Before pursuing a question in another team’s work, ask what connects it to the information, Search, or AI work you can actually observe or influence; what decision it could inform; and what cases or records you can bring. Prefer a question close to observable sales work when that is the relevant consequence—for example, routine pre-sales clarification—before reaching for a revenue claim. Identify who understands the work, who owns the records, and who owns the decision; they may be different people. Bring the observation and its limits to them as an invitation to investigate. Agree what can be inspected and who will decide what, if anything, changes. Continue together while your evidence can help answer the shared question. If the next question belongs elsewhere, carry the established result and remaining uncertainty to the appropriate owner, record their response, and stay available for the next check.

Note: Before you reach out, remember: the person who owns the data may feel measured by it. Frame your inquiry as a request for help solving a shared business uncertainty, not as a data-mining exercise—we are here to support the customer’s best interests.

B6. STATE THE CENTRAL UNCERTAINTY

WE SEE ____________________, BUT WE DO NOT YET KNOW ____________________.

Avoid questions that already contain the answer, such as “How can SEO fix this?” or “What content should we create?”

Take the uncertainty to C.

C — FOLLOW

FOLLOW THE UNCERTAINTY, NOT A PRESET CHAIN

Do not ask, “How far can we trace this?” Ask, “What do we need to understand next for this work or decision?”

Use C whenever the investigation moves.

C1. WHAT DID WE ACTUALLY LEARN?

State the result at the strength established.

C2. GIVEN THAT RESULT, WHAT IMPORTANT THING DO WE NOT KNOW NOW?

C3. WOULD A BETTER ANSWER MATERIALLY CHANGE WHAT WE DO?

This is the governor. More evidence is worth gathering when reducing the remaining uncertainty could materially improve a decision, interpretation, handoff, intervention, or claim enough to justify the cost and delay of learning.

If no: use the result, hand it off, or leave it bounded.

If not sure: name the decision or claim that would use the answer.

C4. WHO OR WHAT IS CLOSEST TO THAT UNCERTAINTY?

A person, a case, a record, a process owner, a data owner, an information environment, a small test, or a specialist.

C5. WHAT IS THE LIGHTEST CREDIBLE NEXT STEP?

One conversation; a few actual cases; a workflow walkthrough; a record review; several independent accounts; an existing operational measure; a short work sample; a comparison of similar cases or periods; an AI re-test; light new instrumentation; specialist input.

“Lightest credible” does not mean “easiest available.” It means the least burdensome step that is still adequate for the claim you need and the consequence of being wrong.

If the next uncertainty sits in another team’s work, continue collaboratively while your evidence can materially help answer the decision. Ask the functional owner to interpret the work and records. If the next step depends on expertise or decision rights you do not own, carry the bounded result and remaining uncertainty to the appropriate owner rather than diagnosing their work or assigning a change.

WHEN THE QUESTION IS ABOUT ORGANIZATIONAL CONSEQUENCE

Start with the people who understand the work. The person who understands the work may not own the data, and the person who owns the data may not understand what the measure means in practice. Treat those as separate jobs until proven otherwise.

Before choosing a measure, learn enough to answer:

What work actually happens?

What is routine, exceptional, or necessary judgment?

What creates waiting, rework, interruption, correction, or escalation?

What does this function spend, protect, preserve, produce, or improve?

What would this team recognize as better or worse?

If people say work fell, where could that work have gone?

Then choose evidence that can distinguish the answers that matter. Useful evidence may include direct accounts, actual cases, tickets, task histories, workflow records, CRM or Support data, product or operational data, short work sampling, comparable cases, direct observation, or light new instrumentation.

MAKE A MEASURE EARN ITS PLACE

A familiar KPI is not automatically the right measure because it is nearby.

Ask:

What does it actually count?

Does it distinguish the kind of work people say changed?

Could it move while the thing we care about stays the same?

Could the thing we care about change while the measure stays flat?

Who understands what is included, excluded, delayed, recoded, or missing?

If total Sales Engineering touches cannot distinguish routine clarification from necessary architecture review, a total-touch count can hide an improvement or create a false efficiency story.

CAN THESE RECORDS SUPPORT THE COMPARISON? Before relying on a record, ask the person who maintains it whether relevant fields were filled consistently, definitions or recording practices changed, entries are stale or duplicated, and missing cases could alter the result. Check a small sample against the underlying work. If a problem remains, record how it limits the comparison and the claim.

CHOOSE EVIDENCE PROPORTIONATE TO THE CLAIM

How strong an answer do you actually need?

LOCAL / BOUNDED UNDERSTANDING — enough for a small operational decision.

BROADER DESCRIPTIVE CLAIM — enough coverage, cases, timing, and denominators where relevant to support a broader statement.

CONTRIBUTION CLAIM — multiple sources, an expected pathway, evidence of exposure where relevant, and important alternative explanations examined.

INCREMENTAL CAUSAL EFFECT — a specialist design capable of estimating what would have happened otherwise.

ECONOMIC / FINANCIAL CLAIM — definitions and assumptions supplied or validated by Finance or the relevant data owner when needed.

Use the lightest method credible enough for the claim you actually need and the consequences of being wrong.

WHAT KIND OF UNCERTAINTY IS THIS?

CALL AI VISIBILITY  Use it when the unresolved question concerns what an AI answer environment finds, retrieves, cites, represents, omits, or recommends.

CALL INFORMATION DESIGN  Use it when the unresolved question concerns what people are trying to accomplish, what information must enable, or where the human-facing environment fails that work.

STAY IN BUSINESS IMPACT  Stay here when the uncertainty concerns organizational work, burden, risk, capacity, customer economics, or another business result.

CALL A SPECIALIST  Use the person or method closest to Product truth, Engineering judgment, Finance definitions, Legal or regulatory interpretation, or formal causal estimation.

When you get an answer, carry forward the result at exactly the strength established. Then take that result to D — LOCATE.

D — LOCATE

LOCATE THE RESULT AND THE CLAIM BOUNDARY

D is not a required sequence of outcomes. Results may enter anywhere, and movement may go in any direction.

RESULT LOCATOR  D can receive and locate results from every territory. Placing a result here does not re-establish a result that another guide or specialist owns.

Use D in two layers.

CORE D: locate the result, mark the claim boundary, write the strongest supported sentence, and choose a disposition.

DEEPER INTERPRETATION: use only when the organizational or economic consequence requires more work.

D1. PLACE THE RESULT

Use four result territories:

INFORMATION ENVIRONMENT — what changed in what information can be found, accessed, retrieved, cited, or represented, whether human-facing or machine-facing?

HUMAN / TASK RESULT — what changed for people trying to accomplish something?

ORGANIZATIONAL CONSEQUENCE — what changed in the organization’s actual work or results?

RESOURCE / ECONOMIC INTERPRETATION — what happened to scarce resources, capacity, cost, risk, margin, throughput, or another economic account?

For every result, record:

CURRENT RESULT

BASIS / SOURCE

EVIDENCE STATE

STILL UNKNOWN

Start with the result you actually have. Leave the other territories blank.

WHERE ELSE CONSEQUENCE CAN LAND

This guide traces consequence into the organization. Sometimes an investigation shows that information-related change lands somewhere else first, or as well. When it does, record it here rather than forcing it into an organizational claim or dropping it.

THE PEOPLE SERVED — customers, members, donors, patients, researchers, residents: fit, successful use, informed non-selection, avoided harm.

THE PEOPLE DOING THE WORK — staff, volunteers, specialists: rework, interruption, protected expert attention.

DELEGATED READERS AND INTERMEDIARIES — AI agents, assistants, or others acting on someone’s behalf: whether a need gets formed, routed, answered, or handed back.

THE WIDER FIELD — peer organizations, shared standards, upstream sources of change.

These are places to look, not impacts to claim. The same evidence states and claim crossings apply. If one becomes the investigation, it may belong with Information Design, AI Visibility, or a specialist, or it may point to work this guide does not yet cover.

D2. THE CLAIM CROSSING

Compare:

WHAT CAN WE CURRENTLY SAY?

WHAT DO WE WANT OR FEEL TEMPTED TO SAY NEXT?

If the second statement requires evidence from another territory, a new uncertainty has appeared.

Forward:

AI answer changed ≠ people received better information.

People performed a task better ≠ Support burden fell.

Support work fell ≠ organizational capacity increased.

Capacity became available ≠ money was saved.

Reverse:

Query volume ≠ demonstrated human need.

Support ticket ≠ information failure.

Poor conversion ≠ content problem.

High specialist workload ≠ missing documentation.

AI misrepresentation ≠ demonstrated business harm.

Organic traffic fell ≠ business value fell.

Organizational result unchanged ≠ no consequence elsewhere

FORWARD QUESTION: if this result matters, what should become observable elsewhere?

BACKWARD QUESTION: if this consequence has an information component, where should evidence of it become visible?

REVERSE INVESTIGATION ≠ REVERSE ATTRIBUTION.

EVERY CROSSING CREATES A NEW UNCERTAINTY.

D3. DISTANCE IS NOT STRENGTH

Keep two questions separate:

HOW FAR DID WE TRACE?

HOW STRONGLY CAN WE SAY IT?

A carefully established close result can be stronger evidence than a distant modeled financial claim. A dollar figure can be precisely calculated while resting on weak causal evidence.

CARRY FORWARD THE RESULT, NOT A STRONGER CLAIM.

DEEPER INTERPRETATION OF ORGANIZATIONAL CONSEQUENCE

A result becomes a useful organizational consequence when you can say what changed, for whom or what, in which cases or conditions, and why that change matters to the operation.

Three different jobs may be needed. You do not need all three for every investigation.

RECOGNITION — someone close to the work can identify a meaningful change and explain what it means in practice.

MEASUREMENT — evidence establishes what changed, how much, how often, for whom, or under what conditions.

VALUATION — the organization estimates or observes what that consequence is worth economically or strategically.

If you are trying to establish a benefit or burden, ask:

What is the result in its native unit?

Who is close enough to the work to recognize whether it matters?

What evidence would show whether it is real beyond one report?

What is the baseline or comparison, if one is needed?

What amount or pattern would count as meaningful for the decision?

What else has to be true for the result to become a realized organizational benefit?

What dependencies could prevent realization even if the close-in result is real?

What disbenefit or burden elsewhere must stay visible?

Who owns the organizational result, even if someone else owns the information change?

Use these questions only when the question requires them. Do not turn a modest investigation into a formal benefits program.

D4. FOLLOW THE WORK OR BURDEN

If workload appears to fall, ask what happened to the work.

It may have been eliminated because the need disappeared; automated; shifted to another role or team; moved earlier or later in the process; deferred; converted into verification or exception handling; absorbed by customers; or hidden until a later failure.

Less work in one place is not automatically less work for the organization.

D5. KEEP OPERATIONAL AND ECONOMIC CLAIMS SEPARATE

Stay in the organization’s native units as long as they are useful. Fewer escalations, less rework, shorter review time, protected specialist attention, lower exception volume, better fit, reduced error, or increased throughput can be meaningful business results without dollar translation.

Do not collapse these claims:

WORKLOAD FELL — less work was required in the observed process or role.

CAPACITY BECAME AVAILABLE — scarce people or systems had usable time or throughput available for something else.

CAPACITY WAS REDEPLOYED — the available capacity was actually used for another valuable purpose.

COST CHANGED — the organization’s cost structure or spend changed.

CASH WAS SAVED — actual cash expenditure was avoided or reduced.

These statements require different evidence.

Time released is not automatically cash saved.

Capacity is not automatically revenue.

Conversion is not automatically profit.

Risk reduced is not automatically loss avoided.

Available capacity can still be a useful result when headcount and cash expense do not change.

IF THE CONSEQUENCE CONCERNS THE CUSTOMER RELATIONSHIP

Keep value created or preserved separate from cost or burden absorbed before collapsing either into a lifetime-value number.

Use the CUSTOMER RELATIONSHIP VALUE & COST MAP in Appendix A to locate what changed and where in the relationship it occurred.

Then use CLV, customer profitability, cost-to-serve, contribution margin, or Finance’s own model only if economic aggregation would improve the decision.

First establish the consequence. Then decide whether it needs valuation.

D6. IF YOU WANT A STRONGER RELATIONSHIP STATEMENT

Describe the relationship at the strength the design supports:

RELATIONSHIP STATEMENT SUPPORTED – NOT EVIDENCE STATE  This ladder describes the kind of relationship statement the design supports. It is separate from the shared evidence-state vocabulary.

DESCRIPTIVE RESULT ONLY — the result was directly recorded in the defined cases or period. This says what happened, not why.

ASSOCIATED WITH — the change and result moved together or differed across a comparison, but important causal alternatives remain.

CONSISTENT WITH THE WORKING EXPLANATION — the pattern fits the expected route and important guardrails held, but the evidence does not justify a causal-contribution claim.

SUPPORTED CONTRIBUTION — the route is plausible, the relevant cases encountered the change, the expected pattern is supported by multiple sources, and important alternative explanations have been examined. This can support a contribution claim without estimating how much of the outcome the intervention caused.

If the decision requires an incremental causal effect — what would have happened otherwise, and by how much — use a design capable of estimating that counterfactual. Experimental, quasi-experimental, or stronger observational methods may be required. That is a specialist-method boundary, not a wording upgrade.

D7. WRITE THE STRONGEST SENTENCE THE EVIDENCE SUPPORTS

Name what changed, for whom, in which cases or period, at what strength, with what important guardrail, and what it does not establish.

Then write:

THE STRONGER SENTENCE WE ARE NOT YET MAKING

and

WHAT WOULD HAVE TO BE ESTABLISHED BEFORE WE COULD MAKE IT.

D8. DO WE NEED TO GO FARTHER?

Would a better answer materially change a real decision?

If yes, take the new uncertainty to C.

If no, the current result may already be sufficient.

If we know enough to act on an information condition, use E.

If information is not the important constraint, use the result, hand it off, or leave it bounded.

WHAT HAPPENS TO THE QUESTION NOW?

Every route ends in a disposition, whether or not anything was changed.

USE THE RESULT — you know enough for the decision in front of you. Act, preserve, scale, revise, or report what the evidence enables.

FOLLOW THE NEXT UNCERTAINTY — another answer would materially change what you should do. Return to C.

HAND IT OFF — the next important question or action belongs with different expertise, ownership, or discipline. Carry the established result, evidence/provenance, important limits, current uncertainty or hypothesis, next domain, and why further work is justified to the appropriate owner. Record their response, any blocker, and the next check. Do not treat the issue as finished solely because it was handed off.

LEAVE IT BOUNDED — you know the present limit and resolving it further is not worth the cost or delay right now. UNKNOWN is a legitimate result.

REFRAME THE INQUIRY — what you learned changed the problem, mechanism, or appropriate entry point. Return to B or C.

NO INFORMATION CHANGE FOR NOW. An investigation can be useful without an information change. Record what the evidence establishes, why a change is not warranted now, and what would justify reopening the question.

E — CHANGE + LEARN

MAKE ONE BOUNDED MOVE AND LEARN FROM IT

E is optional. Not every useful investigation ends with something the practitioner should change.

Use E when the investigation has established enough to justify trying an information change.

LANE GUARD  E is not a substitute for AI Visibility or Information Design. Use E directly when the information condition and proposed change are already clear enough and the learning question is whether that bounded change affects an organizational result. Call the sibling guide when the unresolved question is machine-facing or human-work-facing, then return the resulting evidence to D without changing its strength.

BEFORE THE CHANGE

WHAT DO WE KNOW?

Current result.

Basis / source.

Evidence state.

WHAT INFORMATION CONDITION DO WE THINK WE CAN IMPROVE?

What is missing, unclear, inaccessible, mistimed, fragmented, misleading, inconsistent, hard to carry forward, or otherwise poorly fitted to the work?

Could the important constraint still be something else: product/capability, reliability, UX/workflow, skill, policy, staffing, necessary expert judgment, or organizational process?

DESIGN THE MOVE

WHAT IS THE SMALLEST CHANGE WORTH TRYING?

Small enough to learn from. Substantial enough that a useful difference could appear.

WHO OWNS WHAT?

Who can actually make the change?

Who owns the organizational result we hope to affect, if that is a different person or team?

WHAT SHOULD CHANGE FIRST IF WE ARE RIGHT?

Choose the nearest meaningful result that would actually help decide whether to keep, revise, extend, abandon, or hand off the change.

DECIDE HOW YOU WILL READ IT BEFORE YOU LOOK

WHAT BASELINE OR COMPARISON DO WE NEED?

Current condition, if one is needed.

Meaningful amount or pattern.

Useful comparison: before/after, exposed/unexposed, comparable cases, historical baseline, repeated sample, or none needed for this bounded decision.

BEFORE YOU LOOK, WHAT ELSE COULD HAPPEN?

EXPECTED / SUPPORTIVE.

NO MEANINGFUL CHANGE / NULL.

CONTRADICTORY / COMPLICATING.

ADVERSE / GUARDRAIL FAILURE.

INCONCLUSIVE / INSUFFICIENT EVIDENCE.

WHAT MUST NOT GET WORSE?

When improvement in one place could create a problem elsewhere, choose a balancing measure or other explicit guardrail. Protect warranted scrutiny, necessary friction, safety, and quality.

HOW WILL WE LOOK?

Name the evidence, source, cases or population, timing, and comparison needed. Record important concurrent changes that could materially alter interpretation.

Use the lightest method that can credibly answer the question.

AFTER THE CHANGE

WHAT DID WE ACTUALLY OBSERVE?

Write the observation before the interpretation.

What changed?

What did not?

What happened in the comparison?

What happened to the guardrail?

Which people or cases were represented?

WHAT ELSE CHANGED?

Did anything else happen that could reasonably affect what we observed? Does that materially limit what we can say?

WHAT DO PEOPLE AND RECORDS SAY?

Where do they agree? Where do they conflict?

Treat disagreement as information. A person may notice a local change a dashboard misses; a dashboard may show a broader pattern one person cannot see. Neither automatically overrules the other.

DID THE WORK OR BURDEN MOVE?

Use the D4 possibilities and say what supports the read.

WHICH BEST DESCRIBES THE RESULT?

EXPECTED / SUPPORTIVE — the predicted movement appears under the planned conditions and the guardrail holds.

NO MEANINGFUL CHANGE / NULL — the expected movement does not meaningfully appear under the tested conditions.

INCONCLUSIVE / INSUFFICIENT EVIDENCE — the available evidence or test setup cannot answer the question.

CONTRADICTORY / COMPLICATING — the observations weaken, redirect, or complicate the working explanation.

ADVERSE / GUARDRAIL FAILURE — something important became worse, including a guardrail or burden elsewhere.

WHAT RELATIONSHIP CAN WE SUPPORT?

Use the D6 language if a relationship claim is needed.

WHAT CAN WE NOW SAY?

After we changed ____________________, we observed ____________________.

Write the strongest sentence the evidence supports, the important limits, the stronger statement you are not yet making, and what would have to be established before you could make it.

PUT THE RESULT BACK ON D.

If the organizational consequence now appears to be a real benefit or burden, ask whether another condition must occur before it is actually realized or sustained. Do not confuse a benefit made possible with a benefit actually realized.

Then choose a disposition.

A useful investigation does not have to produce a successful intervention. It has to produce useful learning.

WORKED ROUTE — INDUSTRIAL GATEWAY SECURITY REVIEW

THIRD LEG — ORGANIZATIONAL CONSEQUENCE

CASE SPINE  This teaching case follows one possible route through the family. The guides do not require this order and can be used independently.

This route continues the industrial-gateway teaching case across the field-guide family. It is illustrative, not a client case or a description of a real product. It shows how a bounded human/task result becomes a Business Impact question without changing strength on arrival.

ARRIVE

Under the measured conditions, use and citation of the targeted security evidence increased, product/version representation became more accurate, recommendation did not materially change, and broad category prompts showed little movement.

Under the tested scenarios, security reviewers using the revised reviewer-oriented evidence path identified version applicability more accurately, used fewer disconnected sources, and preserved conditions and authoritative evidence in review more consistently, while appropriate escalation remained stable; task time was inconclusive.

Information Design carries forward:

ESTABLISHED RESULT — the human/task result above, OBSERVED / DOCUMENTED within the tested scenarios.

EVIDENCE / PROVENANCE — comparable-scenario task tests with appropriate reviewers; review notes; escalation records from the tests.

IMPORTANT LIMITS — task time inconclusive; nothing established about live reviews, approval time, Sales Engineering activity, commercial progression, or revenue.

CURRENT HYPOTHESIS — the reduction in reviewer reconstruction reduces routine Sales Engineering clarification during comparable live security reviews.

WHY FURTHER WORK IS JUSTIFIED — the company is deciding whether to extend the governed evidence path to a second product line, and Sales Engineering’s routine-clarification load is part of that decision.

Nothing in that package is upgraded on arrival. The human result stays a human result. The hypothesis is what Business Impact investigates.

SPEC

B1. Doorway: carried-forward evidence from Information Design. Separately, the Sales Engineering lead has said that routine “does it support X?” calls feel fewer. That is a report, recorded as such.

B2. What we are actually seeing: the Information Design result at its stated strength, plus one manager’s account of fewer routine calls. The fact that the account was given is observed; the underlying volume change remains a hypothesis.

The Sales Engineering lead reports that routine clarification calls feel less frequent. We have not yet verified that the volume changed or that the revised evidence path caused it. Could we examine comparable security reviews and separate routine clarification from necessary technical review?

B3. Consequential work: Sales Engineers field pre-sales technical questions from buyer-side security reviewers and implementation engineers. Their work includes routine clarification, necessary architecture review, and packaging evidence a reviewer can carry to Procurement. Good looks like routine clarification handled by the information environment, Sales Engineering attention spent on architecture review and genuinely deployment-specific questions, and reviewers still escalating what warrants escalation.

B4. Information-sized portion: mixed. Better information could plausibly change routine version and connectivity clarification. It would not remove necessary architecture review, deployment-specific verification, entitlement or commercial questions, or unusual configurations. If architecture review fell, that would be a warning rather than an automatic improvement.

B5. Closest to the reality: the Sales Engineering lead, the CRM administrator, and a sample of recent security-review engagements. Permission to inspect the relevant engagement notes is confirmed; individual customers and reviewers are not carried into the readout.

B6. Central uncertainty: WE SEE reviewers in test doing less reconstruction, and the Sales Engineering lead reporting fewer routine calls, BUT WE DO NOT YET KNOW whether routine clarification in live reviews actually declined, whether it moved somewhere else, or whether necessary review held.

FOLLOW

A better answer matters because the organization is deciding whether to extend the evidence path to the second product line. It needs a local, bounded claim. It does not need hours or dollars.

The lightest credible check uses one conversation with the Sales Engineering lead; a review of security-review engagements from two comparable periods, before and after the evidence path went live, with each touch classified as routine clarification, architecture review, or other; a check with Support and the implementation team for the same question class; and a guardrail.

MAKE THE MEASURE EARN ITS PLACE

The obvious measure is total Sales Engineering touches per engagement, which the CRM already reports. The Sales Engineering lead says it would mislead: a routine version question and a two-hour architecture review both count as one touch, and the second kind is work the company wants Sales Engineering doing.

The CRM records touches but not their type.

So the useful measure becomes routine-clarification touches, classified from the engagement notes by someone who understands the work, with architecture-review touches recorded alongside as a guardrail.

The person who understands the work and the person who owns the data are different people. The measure needs both.

The CRM owner confirmed the same engagement inclusion rule and field definitions in both periods, checked duplicate records and note coverage, and found no material difference in missing notes. The Sales Engineering lead checked a sample of classified touches against the original notes; a second reader reviewed a sample of classifications. These checks support a bounded comparison of the recorded engagements, not complete capture of all Sales Engineering work.

CHOOSE EVIDENCE PROPORTIONATE TO THE CLAIM

Two comparable periods, covering all recorded security-review engagements in each, can support a bounded statement about those engagements. They do not support a statement about all Sales Engineering work, total hours, cost, or revenue—and the investigation does not need those claims.

Before reading the result:

SUPPORTIVE — routine version and connectivity clarification declines while architecture-review touches and deployment-specific escalation hold.

NO MEANINGFUL CHANGE — routine clarification does not materially differ.

CONTRADICTORY — routine questions fall in Sales Engineering but appear in Support or as implementation rework.

ADVERSE — deployment-specific escalation falls, suggesting reviewers may be proceeding with false confidence.

TOO AMBIGUOUS — a product release or sales-mix change cannot be separated from the evidence-path change.

LOCATE

For the engagements in the two periods, routine version and connectivity clarification touches decline. Architecture-review touches are stable. Escalation of deployment-specific questions is stable. Support sees no rise in the same question class. The implementation team reports no change in rework attributable to security review. One minor firmware release falls inside the second period; the Sales Engineering lead’s read is that it did not affect the question classes that moved, and that interpretation is recorded as INFERRED.

The Sales Engineering lead and CRM owner also compared the two periods’ security-review engagements by product/version configuration and whether the review involved a new deployment or a repeat of an established configuration. The later period did not shift toward simpler reviews in a way that explains the decline in routine version and connectivity clarification. This comparison addresses the sales-mix alternative identified before the result; it does not isolate the evidence path’s causal effect.

Place the results:

HUMAN / TASK RESULT — established upstream by Information Design and carried here unchanged.

ORGANIZATIONAL CONSEQUENCE — routine Sales Engineering clarification declined in the sampled engagements while necessary review and escalation held. Basis: classified engagement records from two comparable periods plus accounts from Sales Engineering, Support, and implementation. Evidence state: OBSERVED / DOCUMENTED for the sampled engagements. Still unknown: durability beyond the observed periods and whether the pattern holds for the second product line.

RESOURCE / ECONOMIC INTERPRETATION — left blank. No claim is made about Sales Engineering capacity, hours, cost, or value.

THE CLAIM CROSSING

What we can say: routine clarification declined in the sample while necessary review held.

What we are tempted to say: the evidence path freed Sales Engineering capacity.

That crosses into Resource / Economic interpretation. It would require evidence that the released attention was actually available for something else, and—if we wanted a redeployment claim—what it was used for. We did not gather that evidence.

The stronger sentence stays unmade.

RELATIONSHIP CLAIM

The pattern is CONSISTENT WITH THE WORKING EXPLANATION. The firmware-release check and the comparison of product/version configuration and new versus established deployments reduce two named alternatives: the release and a shift toward simpler reviews. The comparison is still two periods, not randomized exposed and unexposed cases, and it cannot establish what share of the decline is attributable to the evidence path. That is enough for the local decision, not for a stronger causal claim.

STRONGEST SENTENCE THE EVIDENCE SUPPORTS

In the security-review engagements sampled from two comparable periods, routine version and connectivity clarification handled by Sales Engineering declined after the reviewer-oriented evidence path was introduced, while architecture-review activity and deployment-specific escalation remained stable and Support saw no corresponding increase. The pattern is consistent with the evidence path reducing routine clarification for the sampled engagements. The evidence does not establish released Sales Engineering capacity, hours, cost, or the share of the change attributable to the path.

STRONGER SENTENCE NOT YET MADE

“The evidence path freed Sales Engineering capacity for higher-value work.”

WHAT WOULD HAVE TO BE ESTABLISHED FIRST

That the released attention became available and what it was actually used for.

DISPOSITION

USE THE RESULT. The organization has what it needs to decide about the second product line, and Sales Engineering can be described narrowly as a beneficiary in the sampled engagements.

The resource/economic question—what any released or redeployed attention is worth—is a new question. Unless the decision requires it, leave it bounded. If Finance does need it, hand off the Readout at exactly this strength.

A second, compact route that enters from an internal customer-relationship signal rather than a carried-forward result appears in Appendix A, A SHORT WORKED SCAN. Reflective Note: At this stage in the real-world investigation, I felt anxious that the Sales Engineering lead was being too casual with their estimates. I had to decide whether to push for ‘harder’ data or use their observation as a thread. I chose the thread. It felt risky, but it was the lightest credible next step—because the intent was to serve the reality of the work, not to push a specific outcome.

APPENDIX A — CUSTOMER RELATIONSHIP ECONOMICS

SEE WHERE A CUSTOMER RELATIONSHIP CREATES VALUE, CONSUMES RESOURCES, AND WHERE INFORMATION MAY PLAUSIBLY AFFECT EITHER

WHAT THIS APPENDIX IS AND IS NOT

This is an optional focused scan within A — SCAN. Use it when you need to find a consequential customer-relationship question, or when mapping the relationship would help frame a problem you already see. Bring what you know about the organization’s customers and the work of serving them; leave with one to three candidate threads, the people and records closest to each, a plausible way information may matter, and what could explain the condition for other reasons. Select one thread and take it to B — SPEC for the investigation. If you already have a specific problem or a result from another guide, you may begin at B without this appendix.

Working with an assistant? See USING THIS GUIDE WITH AN ASSISTANT for the operating rules; the assistant can apply this appendix as the optional focused scan.

It is not a customer lifetime value calculator. It is not a required step in the Business Impact method. It does not ask you to model, forecast, or monetize anything. It draws on established customer-economics practice (customer profitability analysis, cost-to-serve, customer-base analysis, service operations, marketing accountability) and uses that practice’s own language rather than inventing new terms.

Nothing here is a claim.

WHY THE CUSTOMER RELATIONSHIP IS THE TERRAIN

Search practice was selected around one economically important use of information: helping an organization acquire customers. Rankings, traffic, and conversion were convenient places to observe that use. As the click becomes a smaller share of what search does, the observation point weakens. The value information creates does not disappear; it becomes harder to see from where Search was standing.

Conversion is an event in the economic life of a customer, not the definition of success. Before it, a customer has to find, evaluate, and qualify. After it, they have to start, learn, use, get help, decide whether to continue, decide whether to buy more, and eventually leave. Each of those moments can create or preserve value for the organization, and each consumes attention, capacity, or money. Information can plausibly affect either side at almost every moment.

You do not need a sophisticated model to begin seeing this terrain. Customer-base analysis starts by looking at the shape of the base before modeling it, while customer-profitability work keeps value earned and cost absorbed visible together. The map below translates those ideas into a practical orientation tool.

Search does not need to own the customer relationship. It needs to become literate in the economics of the relationships it already influences.

START WITH THIS ORGANIZATION’S ACTUAL ECONOMICS

Do not borrow another industry’s CLV model. Learn where this customer relationship creates value, consumes resources, and creates avoidable burden in this business. Then go meet the people who feel it.

WHAT DOES THIS ORGANIZATION MEAN BY “CUSTOMER”? An account? A subscriber? A buyer? A household? A location? A patient? A student? A donor? A contract? A marketplace participant on one side or both? A buying group of several roles? The answer determines the unit of every row below.

WHAT KIND OF RELATIONSHIP IS IT? First ask whether you can observe when a customer leaves, and whether opportunities to transact occur only at set moments or can occur at any time. Those answers determine which customer states you can observe and which you must estimate.

CONTRACTUAL OR NONCONTRACTUAL. In a contractual relationship (a subscription, a service contract, a membership), the organization knows when a customer leaves; churn is observable and retention can be measured directly. In a noncontractual relationship (retail, ecommerce, most B2B purchasing), the organization does not know whether a quiet customer has left or is simply between purchases; “retention” has to be inferred, and a retention number may not exist.

DISCRETE OR CONTINUOUS. Are purchases or uses discrete events (orders, visits, projects) or a continuous flow (usage, consumption)? This decides whether frequency and recency are the natural measures or whether intensity is.

Typical examples help make the distinction concrete. Many SaaS subscriptions are contractual with continuous usage. Many DTC relationships are noncontractual and discrete. Industrial equipment relationships are often noncontractual and discrete with long gaps, though service contracts may sit alongside them. Higher-education relationships may have explicit term boundaries even when the broader relationship is more complicated. The point is not to force the organization into a quadrant; it is to learn which customer states are actually observable and which must be inferred.

WHAT DOES THIS ORGANIZATION MEAN BY CUSTOMER LIFETIME VALUE, IF IT USES THE TERM? Definitions differ more than the shared name suggests. Ask Finance or the owner of the measure: Is it revenue, gross margin, or contribution? Historical, modeled, or predicted? Discounted? Does it include acquisition cost? Service cost? Over what horizon? What is excluded? Is it a segment average or a per-customer estimate? Do not subtract “customer lifetime costs” from a CLV measure that already includes those costs.

WHERE DO THE NUMBERS ALREADY LIVE? Before doing your own arithmetic, find the organization’s own expressions of customer value and cost. Common locations: investor materials and the annual report for a public company; the board pack; FP&A models; the Customer Success health score and renewal forecast; Support’s cost per contact and contact-reason taxonomy; Sales Operations’ pipeline and win/loss; the returns or refunds report; the implementation team’s hours by account. Each of these is a native unit owned by someone.

WHO OWNS WHAT? The person who understands the work and the person who owns the data are often different people, and for cost-to-serve there is frequently no owner at all. Finding that out is itself a result.

The stable conceptual pair for everything that follows is simple: VALUE CREATED OR PRESERVED BY THE CUSTOMER RELATIONSHIP, and COST OR BURDEN ABSORBED TO ACQUIRE, SERVE, RETAIN, AND GROW IT. The map comes before the formula.

THE ANATOMY — VALUE SIDE

The value side of a customer relationship has a small number of recurring components. Each has a plain meaning, Native units (the language your Finance, Ops, or Support teams already use to count their work), a usual owner, and a caution worth carrying.

ACQUISITION AND FIT. Value begins with a customer who is a good fit for what the organization does. Native units: qualified leads, win rate, fit or disqualification reasons, first-order margin. Usual owners: Marketing, Sales, Sales Operations. Caution: A new customer can create service burden as well as revenue. Check fit and cost-to-serve before treating more acquisition as more value.

MARGIN PER PERIOD. What the relationship earns net of the direct cost of serving it, per month, quarter, or order. Native units: contribution margin per account, gross margin per order, average revenue per user net of cost of revenue. Usual owners: Finance, FP&A. Caution: Revenue alone does not establish margin; ask Finance which direct costs are included.

RETENTION OR SURVIVAL. How long the relationship lasts, and how likely it is to continue. Native units: gross revenue retention, logo churn, renewal rate (contractual); repeat-purchase rate, purchase frequency, time since last purchase (noncontractual). Usual owners: Customer Success, Finance, Marketing analytics. Cautions: A long relationship is not automatically profitable. Customers most likely to leave may not be the ones most responsive to a retention effort.

EXPANSION. Additional purchase, upgrade, cross-buy, or share of wallet from an existing customer. Native units: net revenue retention, expansion revenue, cross-buy rate, share of wallet. Usual owners: Customer Success, Account Management, Sales. Caution: An additional purchase may increase service effort as well as revenue; check the resulting margin.

REFERRAL, INFLUENCE, AND KNOWLEDGE VALUE. Value a customer creates beyond what they buy: referring others, influencing others’ choices, contributing knowledge in reviews, communities, and forums (Kumar et al. 2010; Kumar, Petersen & Leone 2007). Native units: referral rate, review and community contribution, referred-customer value. Usual owners: Marketing, Community, Customer Success. Caution: hard to measure; treat as HYPOTHESIS until shown. This is the value Search practitioners have long stood beside without naming it.

DURATION AND DISCOUNTING. Value that arrives later is worth less than value now, and the length of the relationship compounds everything above. Native units: expected lifetime, discount rate. Usual owner: Finance. Caution: The horizon and discount rate are assumptions. Ask Finance how they were chosen before treating a precise lifetime-value estimate as a measured fact.

The general lesson from the value side: the organization already expresses most of this somewhere, in a native unit, owned by someone. Your job is to find it, not to compute it.

THE ANATOMY — COST SIDE

The cost side is where acquisition-trained practitioners have the least vocabulary and where information most often does its quietest work. Established practice calls these costs customer acquisition cost, cost-to-serve, service cost, retention cost, and, in aggregate, customer profitability. “Lifetime cost” is descriptive language for the same thing.

ACQUISITION COST. What it costs to find, qualify, and win a customer, including the cost of the ones you did not win. Native units: CAC, CAC payback, cost per qualified lead, Sales cycle length, Sales Engineering hours per opportunity. Usual owners: Marketing, Sales, Finance. Caution: Before using an LTV/CAC ratio, check how retention, customer lifetime, acquisition cost, and cost allocation were defined.

ONBOARDING AND IMPLEMENTATION. What it costs to get a customer to first value. Native units: implementation hours, time to value, activation rate, failed first attempts. Usual owners: Implementation, Customer Success, Professional Services. Caution: Ask who records implementation effort, failed first attempts, and time to first value. If no one records them, note that limit.

COST-TO-SERVE. The ongoing cost of Support, service, account management, specialist attention, and correction. Native units: contacts per account, cost per contact, repeat-contact rate, escalation rate, specialist hours by account, service intensity. Usual owners: Support, Customer Success, sometimes nobody. Caution: Service cost may vary substantially across accounts; inspect the distribution before relying on an average. Cost-to-serve is measurable with activity-based methods (Kaplan & Anderson 2007; Guerreiro, Bio & Merschmann 2008).

CONCESSIONS, RETURNS, REFUNDS, AND RECOVERY. The cost of things going wrong: discounts to retain, returns and their handling, refunds, remediation, and the recovery effort after a failure. Native units: return rate, refund rate, concession value, recovery hours. Usual owners: Operations, Finance, Customer Success. Caution: Do not assume fewer returns is always better; examine fit, return policy, margin, and later purchasing together.

CHURN REPLACEMENT. The acquisition cost of replacing a customer who left. Not a separate line in most accounts, but the reason retention matters economically. Native unit: CAC again, multiplied by churn.

RESOURCES SUPPLIED VERSUS RESOURCES USED. The cost side has one distinction that Business Impact’s D5 depends on. Less service work can free capacity without reducing what the organization spends. Record workload, available capacity, redeployment, and cost separately. This is why the map keeps value and cost in native units and resists collapsing them into a dollar figure early.

WHY INFORMATION CAN PLAUSIBLY TOUCH EITHER SIDE

The map’s most important column asks where information might matter. That question needs a plausible mechanism, or it becomes a list of content ideas. Five mechanisms are especially useful.

SEARCH COST AND FIT. Finding and evaluating an option costs the customer effort, and some qualities can only be learned by experience (Stigler 1961; Nelson 1970). Clearer fit information may help people avoid an unsuitable choice. Test whether it changes decisions and downstream work in this setting. When one side knows much more than the other, wrong-fit transactions increase (Akerlof 1970). Information can reduce that asymmetry. This is the mechanism behind FIND / QUALIFY and EXIT / NON-SELECTION.

PURCHASE EASE IN GROUP BUYING. When several people must evaluate the same purchase, ask whether evidence survives the handoff and reduces repeated clarification. This is the mechanism behind BUY / COMMIT.

CUSTOMER-INTRODUCED VARIABILITY. Customers bring variability into service operations: when they arrive, what they ask for, how capable they are, how much effort they will make, and what they prefer. Information may help customers state what they need or perform a task they can reasonably do themselves; it cannot remove every source of variation or necessary service. This is the mechanism behind ONBOARD, USE, and SUPPORT, and it also tells you what would remain difficult for other reasons.

SELF-SERVICE AND EFFORT. Self-service that works is associated with retention, but the effect depends on the mix with personal service (Scherer, Wünderlich & von Wangenheim 2015; Meuter et al. 2000), and the strongest service driver of loyalty is low customer effort rather than delight (Dixon, Freeman & Toman 2010). If customers can resolve a routine question successfully, contact demand may fall. Check task success and whether work moved to another channel before claiming lower cost.

CUSTOMER EDUCATION. Education can improve understanding while also helping customers evaluate alternatives. Whether that benefits this organization is an empirical question. This is why USE / SUCCEED and RETAIN / RENEW are both places information matters and places where the effect is not guaranteed to favor the organization.

And one boundary that applies to all five: information changes what customers can know and ask for. It does not change product capability, reliability, entitlement, policy, staffing, or the judgment a task genuinely requires. Every row of the map carries the question: what would still remain for other reasons?

MAP THE ECONOMIC LIFE OF THIS CUSTOMER

Use the relationship areas below as prompts, not as a universal lifecycle. Remove rows that do not fit. Add the ones this organization actually has. The row names are the same ones Business Impact uses in its own customer-relationship table, compressed there and complete here.

CONTINUED — MEASURES, INFORMATION MECHANISM, AND OTHER CONDITIONS

For each meaningful row, ask, in order:

WHAT COULD BETTER INFORMATION PLAUSIBLY CHANGE HERE? (Which of the five mechanisms in WHY INFORMATION CAN PLAUSIBLY TOUCH EITHER SIDE is in play?)

WHAT WOULD STILL REMAIN FOR OTHER REASONS?

WHO INSIDE THE ORGANIZATION IS CLOSEST TO THE CONSEQUENCE, AND WHO OWNS THE NUMBER?

WHAT EVIDENCE WOULD TELL US WHETHER THIS IS REAL?

The mechanisms most often relevant by row: FIND / QUALIFY and EXIT, search cost and fit; BUY / COMMIT, purchase ease in group buying; ONBOARD, USE, and SUPPORT, customer-introduced variability and self-service and effort; USE and RETAIN, customer education; EXPAND, customer education and knowledge value. These are the usual suspects, not a rule.

DIFFERENT BUSINESSES, DIFFERENT ANATOMIES

Do not borrow another industry’s customer lifecycle or CLV anatomy. The sketches below show how the same ledger takes different shapes. Each names the relationship type, where value develops, where cost accumulates, the native measures and their usual owners, where information often matters, and the caution the type most needs.

ECOMMERCE / DIRECT-TO-CONSUMER. Noncontractual, discrete. Value develops through orders, gross margin, repeat purchase, frequency, and lower return rate. Cost accumulates through acquisition, fulfillment problems, returns, refunds, Support, and failed-fit demand. Native measures: repeat rate, purchase frequency, average order value, return rate, contribution margin per order (Marketing, Merchandising, Operations, Finance). Information often matters around fit, sizing and specification, expectation, comparison, use, troubleshooting, return prevention, and repeat discovery. Caution: in a noncontractual relationship, “retention” is inferred, not observed; a quiet customer may not have left. Returns are not simply a cost to suppress.

SAAS / SUBSCRIPTION. Contractual, continuous. Value develops through activation, recurring margin, retention, successful use, and expansion. Cost accumulates through acquisition, implementation, onboarding, Support, Customer Success, concessions, and churn replacement. Native measures: net and gross revenue retention, logo churn, CAC payback, activation and adoption, time to value, health score (Customer Success, Finance, Product). Information often matters around qualification, setup, time to value, capability discovery, adoption, troubleshooting, renewal evidence, and appropriate expansion. Caution: benchmark figures for these measures should come from primary surveys, not from vendor pages. Confirm which measures and decisions Customer Success, Product, and Finance actually own here.

COMPLEX B2B / INDUSTRIAL. Noncontractual, discrete, with long gaps and few large accounts. Value develops through qualified accounts, account margin, successful deployment, productive operation, retention, and expansion or replacement. Cost accumulates through long Sales cycles, Sales Engineering, specialist review, procurement work, implementation, field support, and troubleshooting. Native measures: account profitability, Sales Engineering hours by opportunity and by type, cycle length, deployment success, service intensity (Sales, Sales Engineering, Finance, Service). Information often matters around applicability, limitations, technical evidence, the reviewer-to-Procurement handoff, prerequisites, configuration, governed product truth, and preserving expert attention for warranted scrutiny. Caution: the customer is a buying center of several roles with different information requirements.

PROFESSIONAL SERVICES. Contractual by engagement, discrete. Value develops through qualified engagements, utilization, repeat work, referral, and profitable delivery. Cost accumulates through selling, scoping, delivery effort, rework, scope creep, and expert involvement before fit is known. Native measures: utilization, realization, repeat and referral rate, scope change (Partners, Operations, Finance). Information often matters around self-qualification, expectation setting, scoping evidence, preparation, process guidance, and the handoff into delivery. Caution: the economic anatomy varies substantially by firm. Start with the organization’s own scoping, utilization, realization, and rework definitions.

REGULATED / HIGH-STAKES WORK. Relationship type varies. Value develops through appropriate use, successful outcomes, lower preventable error, durable relationships, and preserved trust. Cost accumulates through specialist review, compliance work, preventable mistakes, rework, unnecessary escalation, and risk exposure. Native measures: error and exception rates, review time, escalation rate, compliance findings (Compliance, Quality, Legal, Operations). Information often matters around correct routing, provenance, applicability, limits, condition recognition, and preserving necessary expert review. Caution: less friction is not automatically better; some verification, review, and escalation exist because an obligation or necessary judgment cannot disappear.

MARKETPLACES / PLATFORMS. Two-sided; often noncontractual on one side and contractual on the other. Value develops through liquidity, transactions, take rate, and retention on both sides. Cost accumulates through acquisition on both sides, trust and safety, disputes, and Support for participants who are also each other’s customers. Native measures: take rate, gross merchandise value, liquidity, repeat rate by side, dispute rate (Growth, Operations, Finance). Information often matters around fit between sides, expectation accuracy, dispute prevention, and self-service for participants.  Caution: not all marketplaces are alike; the economics depend on frequency and fragmentation.

MEMBERSHIP, NONPROFIT, AND DONOR RELATIONSHIPS. Contractual or quasi-contractual, often annual. Value develops through retained members or donors, upgrade, referral, and mission outcomes. Cost accumulates through acquisition, stewardship, and lapsed-donor replacement. Native measures may include member or donor retention, average gift, lifetime giving, and cost to raise a dollar (Development, Finance). Information often matters around expectation accuracy, impact evidence, and the ease of continuing. Caution: use the organization’s own mission and fundraising economics rather than importing a commercial lifetime-value formula.

MISSION-DRIVEN ENROLLMENT (HIGHER EDUCATION AND SIMILAR). Contractual by term. Value develops through enrollment, persistence, completion, and outcomes; expectation accuracy at entry drives everything after. Cost accumulates through recruitment, advising, support services, and attrition. Native measures: yield, persistence and completion rates, cost per completion, time to degree (Enrollment, Student Success, Institutional Research). Information often matters around fit, requirements, realistic expectations, access to support, and informed non-enrollment. Caution: the institution’s own persistence, completion, support, and outcome economics are more useful than importing a generic lifetime-value formula. More enrollment is not a sufficient objective function.

A CROSS-CUTTING NOTE: THE CUSTOMER AS A WORKING SYSTEM. In many of the types above, and especially in regulated B2B services, the “customer” is not one person moving through a funnel but several roles with different information requirements across selection and delivery: clinical operations, procurement, regulatory, quality, data, finance. Treat any internal model of that system as a hypothesis until real cases corroborate it. Map the roles before you map the lifecycle.

THREE WAYS IN

The map can be entered from any direction. If you do not yet have a thread, use this focused scan to find one. If you already have a specific problem or a result carried forward from another guide, you can go directly to B — SPEC; use this appendix only when mapping the customer relationship would help frame the question.

FROM A SEARCH OR AI SIGNAL, OUTWARD. You have a query pattern, an AI answer that misrepresents or omits something, a traffic change, or a content request. First move: name what the signal actually establishes (people searched; the system represented; the measure moved) and nothing more. Second move: find the row. Which part of the relationship is this signal near? Third move: ask who feels it inside the organization and what they already measure. Standing rule: a query is not a demonstrated need; an AI misrepresentation is not demonstrated harm; traffic fell is not value fell. Take the thread to B.

FROM AN INTERNAL RELATIONSHIP PROBLEM, BACKWARD. You have been handed, or noticed, a cost or value problem: Support volume in one question class, Sales Engineering load, churn in a segment, returns, slow onboarding, wrong-fit demand. First move: understand the work in its native unit before touching a measure; find the person who understands it and the person who owns the data. Second move: ask which of the five information mechanisms could plausibly be in play, and what would remain for other reasons. Third move: find where evidence of an information component would become visible if it existed. Standing rule: a costly problem does not establish information as the cause; reverse investigation is not reverse attribution. Take the thread to B.

FROM THE CUSTOMER BASE, SIDEWAYS. You have access to transaction, account, or engagement data and want to look before choosing. First move: place the organization in the relationship-type grid so you know which measures can exist. Second move: review the distribution of value, service burden, and cohort behavior before choosing a thread. Third move: for any pattern that catches your attention, find the row and the owner. Standing rule: a pattern in the base is a place to look, not a claim about why it is there. Take the thread to B.

BRING INFORMATION DESIGN AND AI VISIBILITY ONTO THE MAP

For any relationship area that matters, two questions open the specific investigations the sibling guides own.

THE INFORMATION DESIGN QUESTION. What are people trying to accomplish here? What must information enable them to understand, judge, verify, do, or pass onward? Where does the current information environment make the work harder, less reliable, more expensive, or more dependent on scarce expertise? What must remain attached to evidence as it moves between people?

THE AI VISIBILITY QUESTION. Where might AI systems now mediate the same information environment: discovery, comparison, fit, limitations, prerequisites, technical evidence, troubleshooting, value evidence, capability discovery? What do they retrieve, cite, omit, represent, or recommend about this part of the relationship? Is that behavior a research lead about a human need, or only a fact about the machine?

Two limits to carry. Information Design and AI Visibility can change what people and systems can find, understand, verify, and carry. They cannot change product capability, reliability, entitlement, policy, staffing, or necessary judgment. And a plausible information mechanism is not evidence of economic impact; it is a reason to investigate. The output of the map is one or more places worth investigating.

MORE IS NOT AUTOMATICALLY BETTER

Customer economics repeatedly punishes simple optimization stories. Keep these provocations visible while you scan:

MORE ACQUISITION can mean more wrong-fit customers.

HIGHER CONVERSION can mean fewer people correctly self-selecting out.

LONGER RETENTION can preserve valuable relationships — or prolong unprofitable ones.

MORE EXPANSION can create margin — or costly, inappropriate cross-buying.

LOWER SUPPORT VOLUME can mean better self-service — or customers giving up, moving work elsewhere, or failing later.

LESS EXPERT INVOLVEMENT can mean protected scarce attention — or lost warranted scrutiny.

MORE CUSTOMER EDUCATION can increase trust and capability — and can also make customers better able to evaluate alternatives.

The question is not whether the number moved in the preferred direction. The question is what “better” means in this relationship, in the organization’s own units, under the conditions that matter.

A channel report is a doorway, not a result.

FORAGE INTERNALLY BEFORE YOU MODEL

Once a row and a mechanism have caught your attention, the next step is to get closer to the organization’s own evidence, lightly and with permission.

THE STANCE. Look before you model. Ask what the organization already knows, in the units it already uses, owned by the people who already own it.

THE DOCUMENTS TO ASK FOR. The board pack or investor materials; the FP&A model’s customer assumptions; the Customer Success health score definition and renewal forecast; Support’s contact-reason taxonomy and cost per contact; the returns or refunds report; Sales Operations’ win/loss and disqualification reasons; the implementation team’s hours by account; any existing customer profitability or cost-to-serve analysis, however old.

THE PEOPLE TO MEET. The person who understands the work in the row you chose. The person who owns the data about it. The person who owns the organizational result you hope to affect, if different. One conversation with each is usually the lightest credible first step.

THE QUESTIONS. What work actually happens here? What is routine, what is exceptional, what requires judgment? What creates waiting, rework, interruption, or escalation? What does this function spend, protect, produce, or improve? What would you recognize as better or worse? If this work fell, where would it go? What does your measure actually count, and what does it miss?

WORKING WITH FIRST-PARTY RECORDS. Before opening tickets, CRM records, call notes, task histories, or account data, confirm that you have permission to use them for this purpose. Prefer patterns to individuals. Anonymize accounts. Do not carry individual-level detail into a readout. Respect the client’s or organization’s data terms. If the evidence cannot be obtained appropriately, record UNKNOWN.

WHAT THE AUDIT STANCE MISSES. Customer-base analysis assumes transaction data. In many organizations the most important customer facts live in Sales Engineering notes, Support tickets, implementation records, and people’s heads. The map is designed to find those, too.

PICK A THREAD

For each candidate, record:

OBSERVED QUESTION:

NEAREST OWNER / SOURCE:

POSSIBLE INFORMATION MECHANISM:

COMPETING EXPLANATION:

NEXT UNCERTAINTY:

Choose one to three. Pick one that is real enough to investigate, consequential enough to care about, and close enough that you can learn something.

Take it to B — SPEC. From here, Business Impact’s method does the work; the map has done its job.

A SHORT WORKED SCAN

Illustrative, not a client case.

An in-house SEO at a mid-sized SaaS company has watched organic sessions fall for a year and cannot show what was lost. She uses the map sideways rather than forward: instead of starting from the traffic, she asks Customer Success what they hear on renewal calls. The CS lead says the same thing keeps surfacing: “We didn’t know it could do that.” Customers renewing at the base tier are discovering, at renewal, capabilities they have been paying for and not using; some downgrade, a few churn, and the CS team spends the renewal conversation teaching rather than expanding.

She places the organization: contractual, continuous. She finds the row: USE / SUCCEED, with a foot in RETAIN / RENEW and EXPAND / REPEAT. The native units already exist: adoption by feature (Product analytics), renewal outcome and downgrade rate by cohort (Customer Success, Finance), and CS hours per renewal (nobody owns this; the CS lead estimates). The mechanism is customer education and capability discovery, with self-service and effort behind it. What would remain for other reasons: some customers bought the wrong tier (fit), some features do not work for their use case (capability), some will downgrade regardless (price).

The Information Design question: what are base-tier customers trying to accomplish in months three through ten, and what must information enable them to discover and try without a call? The AI Visibility question: when those customers ask an assistant “can [product] do X,” what does it retrieve and represent, and does it send them to a capability they already own or to a competitor?

Her thread: WE SEE renewal calls repeatedly surfacing unknown capabilities among base-tier customers, and downgrades in that cohort, BUT WE DO NOT YET KNOW how prevalent the discovery gap is, whether it drives the downgrades, or whether the information environment (in-product, help center, or AI-mediated) is where the gap lives. Who can get her closer: the CS lead, a sample of renewal-call notes with permission, Product’s adoption data. She takes it to B.

Nothing has been claimed. A place has been found, in the organization’s own units, with owners attached, and a reason to think information might matter that is more than a wish.

For the research and practitioner lineage behind this focused scan, see CUSTOMER ECONOMICS AND MARKETING ACCOUNTABILITY in METHODS, LINEAGE, AND SOURCES.

APPENDIX B — THE BUSINESS IMPACT READOUT

Use one Readout per bounded investigation. Build it as the investigation develops. Not every field is required. UNKNOWN is a legitimate entry.

MINIMUM USEFUL RECORD

For a modest question, this is enough:

Use the fuller record when stakes, multiple people, contested measures, or a stronger claim make the extra discipline useful.

Need more structural rigor? Continue into the full Readout.

FULL BUSINESS IMPACT READOUT

ARRIVAL

SPEC

WE SEE ____________________, BUT WE DO NOT YET KNOW ____________________.

FOLLOW LOG

For each important move, record:

CONSEQUENCE SKETCH

INFORMATION ENVIRONMENT

HUMAN / TASK RESULT

ORGANIZATIONAL CONSEQUENCE

RESOURCE / ECONOMIC INTERPRETATION

CONSEQUENCE NOTED OUTSIDE THE ORGANIZATION — OPTIONAL

STRONGEST SENTENCE THE EVIDENCE SUPPORTS

STRONGER STATEMENT WE ARE NOT YET MAKING

CHANGE + LEARN, IF E WAS USED

DISPOSITION

EVIDENCE TO CARRY FORWARD — IF NEEDED

HANDOFF RULE  A handoff begins a new question. It does not add another conclusion to the originating investigation. Copy the established result without rewriting it; enter the hypothesis as an uncertainty to investigate.

METHODS, LINEAGE, AND SOURCES

This guide adapts established approaches to a particular question: what can we responsibly say about the organizational consequence of an information condition or change? The approaches below supply different parts of that work. They do not turn a Search, Content, or AI practitioner into the owner of another team’s records, an accountant, or a causal-inference specialist. Use the part that helps answer the decision in front of you, carry its limits forward, and involve the person or method closest to a question you cannot establish here.

HOW THESE METHODS INFORM THE GUIDE

PROGRAM AND PROCESS EVALUATION. Helps frame a decision-worthy question, understand the setting, gather evidence that fits the question, and use a bounded finding. In this guide it informs B’s inquiry specification, C’s evidence choices, and D’s disposition. It does not make every local investigation a formal program evaluation.

THEORY-BASED EVALUATION AND CONTRIBUTION ANALYSIS. Helps state a plausible pathway from an information condition to observed work, specify what should become visible if the pathway matters, and examine rival explanations. The guide uses this discipline for working explanations and cautious contribution language in C and D. A plausible pathway is not an estimated causal effect.

BENEFITS REALIZATION MANAGEMENT. Helps ask whether a close result actually became a benefit, who owns that benefit, what dependencies or disbenefits matter, and whether it lasted. D applies these distinctions before crossing from work or capacity into a claimed organizational or economic gain.

IMPROVEMENT SCIENCE. Helps make a bounded change, decide in advance what would support or weaken the expectation, and watch a guardrail so improvement in one place does not conceal harm elsewhere. E uses this approach only when a change is warranted.

DECISION ANALYSIS AND VALUE OF INFORMATION. Helps decide whether another answer is worth its cost and delay. C asks whether reducing the next uncertainty could materially change the decision; otherwise the result can be used, handed over, or left bounded.

MANAGERIAL ACCOUNTING AND CAPACITY ANALYSIS. Helps separate less work from capacity made available, its actual redeployment, a changed cost, and cash saved. D keeps these as separate claims and calls on Finance or the relevant owner for definitions and financial interpretation.

CUSTOMER ECONOMICS AND MARKETING ACCOUNTABILITY. Helps see where a customer relationship creates or preserves value and consumes resources. Appendix A adapts these ideas as an optional focused scan for a thread; it is not a CLV calculator and does not establish business impact before B through D do their work.

SEARCH AND AI PRACTICE. Search signals, AI answer observations, and information interventions can bring a practitioner to this method. Citation Labs’ precedents explain that entry, while the other disciplines govern how far an organizational claim can travel. Click behavior is an observation condition, not by itself a measure of business value.

The sources below are grouped by the work they inform. A link beside a specific claim in the guide supports that claim at the point of use; the grouped list lets readers inspect the wider method and its lineage. We have adapted selected ideas, not adopted every model or conclusion from every source. Where a decision requires formal estimation, accounting, legal, regulatory, or domain judgment, take the bounded question to the appropriate specialist.

SOURCES

Selected sources for the method are grouped by lineage. Customer Relationship Economics sources appear at the end of Appendix A.

PROGRAM AND PROCESS EVALUATION

Kidder, D. P., et al. (2024). CDC Program Evaluation Framework, 2024. MMWR Recommendations and Reports, 73(6). source

Patton, M. Q. (2008). Utilization-Focused Evaluation (4th ed.). Sage. source

THEORY-BASED EVALUATION AND CONTRIBUTION ANALYSIS

Weiss, C. H. (1995). Nothing as practical as good theory: Exploring theory-based evaluation for comprehensive community initiatives for children and families. In Connell, J. P., et al. (Eds.), New Approaches to Evaluating Community Initiatives. Aspen Institute. source

Funnell, S. C., & Rogers, P. J. (2011). Purposeful Program Theory: Effective Use of Theories of Change and Logic Models. Jossey-Bass. source

Mayne, J. (2012). Contribution analysis: Coming of age? Evaluation, 18(3). source

BENEFITS REALIZATION MANAGEMENT

Bradley, G. (2010). Benefit Realisation Management: A Practical Guide to Achieving Benefits Through Change (2nd ed.). Gower. source

Ward, J., & Daniel, E. (2012). Benefits Management: How to Increase the Business Value of Your IT Projects (2nd ed.). Wiley. source

Project Management Institute. (2019). Benefits Realization Management: A Practice Guide. PMI. source

IMPROVEMENT SCIENCE

Langley, G. J., Moen, R. D., Nolan, K. M., Nolan, T. W., Norman, C. L., & Provost, L. P. (2009). The Improvement Guide: A Practical Approach to Enhancing Organizational Performance (2nd ed.). Jossey-Bass. source

Provost, L. P., & Murray, S. K. (2011). The Health Care Data Guide: Learning from Data for Improvement. Jossey-Bass. source

DECISION ANALYSIS AND VALUE OF INFORMATION

Howard, R. A. (1966). Information value theory. IEEE Transactions on Systems Science and Cybernetics, SSC-2(1), 22–26. source

Raiffa, H. (1968). Decision Analysis: Introductory Lectures on Choices under Uncertainty. Addison-Wesley. source

Hubbard, D. W. (2014). How to Measure Anything: Finding the Value of Intangibles in Business (3rd ed.). Wiley. source

MANAGERIAL ACCOUNTING AND RESOURCE-CAPACITY ANALYSIS

Cooper, R., & Kaplan, R. S. (1991). Profit priorities from activity-based costing. Harvard Business Review, May–June. source

Cooper, R., & Kaplan, R. S. (1992). Activity-based systems: Measuring the costs of resource usage. Accounting Horizons, 6(3), 1–13. source

Kaplan, R. S., & Anderson, S. R. (2007). Time-Driven Activity-Based Costing: A Simpler and More Powerful Path to Higher Profits. Harvard Business School Press. source

Kaplan, R. S., & Narayanan, V. G. (2001). Measuring and managing customer profitability. Journal of Cost Management, 15(5), 5–15. source

CUSTOMER ECONOMICS AND MARKETING ACCOUNTABILITY

Gupta, S., & Lehmann, D. R. (2005). Managing Customers as Investments: The Strategic Value of Customers in the Long Run. Wharton School Publishing. source

Rust, R. T., Lemon, K. N., & Zeithaml, V. A. (2004). Return on marketing: Using customer equity to focus marketing strategy. Journal of Marketing, 68(1), 109–127. source

Ambler, T. (2003). Marketing and the Bottom Line: The Marketing Metrics to Pump Up Cash Flow (2nd ed.). FT Prentice Hall. source

Hanssens, D. M., & Pauwels, K. H. (2016). Demonstrating the value of marketing. Journal of Marketing, 80(6), 173–190. source

Marketing Accountability Standards Board. Marketing Metric Audit Protocol. source

RELATIONSHIP TYPES AND CUSTOMER ECONOMICS

Wübben, M., & von Wangenheim, F. (2008). Instant customer base analysis: Managerial heuristics often “get it right.” Journal of Marketing, 72(3). source

Fader, P., Hardie, B. G. S., & Ross, M. (2022). The Customer-Base Audit: The First Step on the Journey to Customer Centricity. Wharton School Press. source

Gupta, S., Hanssens, D., Hardie, B., Kahn, W., Kumar, V., Lin, N., Ravishanker, N., & Sriram, S. (2006). Modeling customer lifetime value. Journal of Service Research, 9(2). source

Kumar, V., & Reinartz, W. (2016). Creating enduring customer value. Journal of Marketing, 80(6). source

Kumar, V. (2018). A theory of customer valuation. Journal of Marketing, 82(1). source

Berger, P. D., & Nasr, N. I. (1998). Customer lifetime value: Marketing models and applications. Journal of Interactive Marketing, 12(1), 17–30. source

Peppers, D., & Rogers, M. (2005). Return on Customer: Creating Maximum Value from Your Scarcest Resource. Doubleday. source

Fader, P. S., & Hardie, B. G. S. (2009). Probability models for customer-base analysis. Journal of Interactive Marketing, 23(1). source

McCarthy, D. M., & Fader, P. S. (2018). Customer-based corporate valuation for publicly traded noncontractual firms. Journal of Marketing Research, 55(5). source

McCarthy, D. M., Fader, P. S., & Hardie, B. G. S. (2017). Valuing subscription-based businesses using publicly disclosed customer data. Journal of Marketing, 81(1). source

Kumar, V., & Shah, D. (2009). Expanding the role of marketing: From customer equity to market capitalization. Journal of Marketing, 73(6). source

Schulze, C., Skiera, B., & Wiesel, T. (2012). Linking customer and financial metrics to shareholder value. Journal of Marketing, 76(2). source

Gupta, S., Lehmann, D. R., & Stuart, J. A. (2004). Valuing customers. Journal of Marketing Research, 41(1), 7–18. source

VALUE SIDE

Reinartz, W., & Kumar, V. (2002). The mismanagement of customer loyalty. Harvard Business Review, July. source

Reinartz, W., & Kumar, V. (2000). On the profitability of long-life customers in a noncontractual setting: An empirical investigation and implications for marketing. Journal of Marketing, 64(4), 17–35. source

Bolton, R. N. (1998). A dynamic model of the duration of the customer’s relationship with a continuous service provider. Marketing Science, 17(1). source

Ascarza, E. (2018). Retention futility: Targeting high-risk customers might be ineffective. Journal of Marketing Research, 55(1). source

Keiningham, T. L., Cooil, B., Andreassen, T. W., & Aksoy, L. (2007). A longitudinal examination of Net Promoter and firm revenue growth. Journal of Marketing, 71(3). source

Shah, D., Kumar, V., Qu, Y., & Chen, S. (2012). Unprofitable cross-buying: Evidence from consumer and business markets. Journal of Marketing, 76(3). source

Kumar, V., Aksoy, L., Donkers, B., Venkatesan, R., Wiesel, T., & Tillmanns, S. (2010). Undervalued or overvalued customers: Capturing total customer engagement value. Journal of Service Research, 13(3). source

Kumar, V., Petersen, J. A., & Leone, R. P. (2007). How valuable is word of mouth? Harvard Business Review, October. source

Zeithaml, V. A., Rust, R. T., & Lemon, K. N. (2001). The customer pyramid: Creating and serving profitable customers. California Management Review, 43(4). source

Mittal, V., Sarkees, M., & Murshed, F. (2008). The right way to manage unprofitable customers. Harvard Business Review, April. source

COST SIDE

Shapiro, B. P., Rangan, V. K., Moriarty, R. T., & Ross, E. B. (1987). Manage customers for profits (not just sales). Harvard Business Review, September. source

Homburg, C., Droll, M., & Totzek, D. (2008). Customer prioritization: Does it pay off, and how should it be implemented? Journal of Marketing, 72(5). source

Guerreiro, R., Bio, S. R., & Merschmann, E. V. V. (2008). Cost-to-serve measurement and customer profitability analysis. International Journal of Logistics Management, 19(3). source

Niraj, R., Gupta, M., & Narasimhan, C. (2001). Customer profitability in a supply chain. Journal of Marketing, 65(3), 1–16. source

Petersen, J. A., & Kumar, V. (2009). Are product returns a necessary evil? Antecedents and consequences. Journal of Marketing, 73(3). source

Hochstein, B., Rangarajan, D., Mehta, N., & Kocher, D. (2020). An industry/academic perspective on customer success management. Journal of Service Research, 23(1). source

Ryals, L. (2005). Making customer relationship management work: The measurement and profitable management of customer relationships. Journal of Marketing, 69(4). source

INFORMATION MECHANISMS

Stigler, G. J. (1961). The economics of information. Journal of Political Economy, 69(3), 213–225. source

Nelson, P. (1970). Information and consumer behavior. Journal of Political Economy, 78(2), 311–329. source

Akerlof, G. A. (1970). The market for “lemons”: Quality uncertainty and the market mechanism. Quarterly Journal of Economics, 84(3), 488–500. source

Toman, N., Adamson, B., & Gomez, C. (2017). The new sales imperative. Harvard Business Review, March–April. source

Frei, F. X. (2006). Breaking the trade-off between efficiency and service. Harvard Business Review, November. source

Frei, F. X., & Morriss, A. (2012). Uncommon Service: How to Win by Putting Customers at the Core of Your Business. Harvard Business Review Press. source

Meuter, M. L., Ostrom, A. L., Roundtree, R. I., & Bitner, M. J. (2000). Self-service technologies: Understanding customer satisfaction with technology-based service encounters. Journal of Marketing, 64(3). source

Scherer, A., Wünderlich, N. V., & von Wangenheim, F. (2015). The value of self-service: Long-term effects of technology-based self-service usage on customer retention. MIS Quarterly, 39(1). source

Dixon, M., Freeman, K., & Toman, N. (2010). Stop trying to delight your customers. Harvard Business Review, July–August. source

Eisingerich, A. B., & Bell, S. J. (2008). Customer education increases trust. MIT Sloan Management Review, 50(1). source

Bell, S. J., & Eisingerich, A. B. (2007). The paradox of customer education: Customer expertise and loyalty in the financial services industry. European Journal of Marketing, 41(5/6), 466–486. source

Bell, S. J., Auh, S., & Eisingerich, A. B. (2017). Unraveling the customer education paradox: When, and how, should firms educate their customers? Journal of Service Research, 20(3). source

DIFFERENT BUSINESSES

KeyBanc Capital Markets & Sapphire Ventures. (2025). SaaS Survey. source

High Alpha. (2025). SaaS Benchmarks Report (formerly OpenView). source

Skok, D. SaaS Metrics 2.0: A guide to measuring and improving what matters. forEntrepreneurs. source

Venkatesan, R., & Kumar, V. (2004). A customer lifetime value framework for customer selection and resource allocation strategy. Journal of Marketing, 68(4). source

Webster, F. E., & Wind, Y. (1972). A general model for understanding organizational buying behavior. Journal of Marketing, 36(2), 12–19. source

Gartner. The B2B buying journey: Key stages and how to optimize them. source

Maister, D. H. (1993). Managing the Professional Service Firm. Free Press. source

Kotler, P., & Levy, S. J. (1971). Demarketing, yes, demarketing. Harvard Business Review, November–December. source

Gurley, B. (2012, November 13). All markets are not created equal: 10 factors to consider when evaluating digital marketplaces. Above the Crowd. source

Sargeant, A. (2001). Using donor lifetime value to inform fundraising strategy. Nonprofit Management and Leadership, 12(1). source

Pausits, A. Student lifetime value: A paradigm shift from “drive-through university” to partner for life. source

Lemon, K. N., & Verhoef, P. C. (2016). Understanding customer experience throughout the customer journey. Journal of Marketing, 80(6). source

CAUTIONS

Gurley, B. (2012). The dangerous seduction of the lifetime value (LTV) formula. Above the Crowd. source

Lewis, R. A., & Rao, J. M. (2015). The unfavorable economics of measuring the returns to advertising. Quarterly Journal of Economics, 130(4). source

Blake, T., Nosko, C., & Tadelis, S. (2015). Consumer heterogeneity and paid search effectiveness: A large-scale field experiment. Econometrica, 83(1). source

Gordon, B. R., Zettelmeyer, F., Bhargava, N., & Chapsky, D. (2019). A comparison of approaches to advertising measurement: Evidence from big field experiments at Facebook. Marketing Science, 38(2). Source

CITATION LABS PRECEDENTS

French, G., Wirth, J., & Arnett, M. (2026, February 20). Yes, AI rank is trackable: Results from 4,579 buyer prompts. Citation Labs. source

French, G., & Wirth, J. (2026, March 13). Measuring 3rd-party influence on AI answers. Citation Labs. source

THE CLICK AS OBSERVATION POINT SparkToro. (2026). In 2026, less than one third of Google searches still send a click. https://sparktoro.com/blog/in-2026-less-than-one-third-of-google-searches-still-send-a-click/ Pew Research Center. (2025, July 22). Google users are less likely to click on links when an AI summary appears in the results. https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/ Indig, K., & Johnson, A. The great decoupling. Growth Memo. https://www.growth-memo.com/p/the-great-decoupling

French, G., Wirth, J., & Arnett, M. (2026, February 20). Yes, AI rank is trackable: Results from 4,579 buyer prompts. Citation Labs. source

French, G., & Wirth, J. (2026, March 13). Measuring 3rd-party influence on AI answers. Citation Labs. source

THE CLICK AS AN OBSERVATION POINT

SparkToro. (2026). In 2026, less than one third of Google searches still send a click. source

Pew Research Center. (2025, July 22). Google users are less likely to click on links when an AI summary appears in the results. source

Indig, K., & Johnson, A. The great decoupling. Growth Memo. source

WHAT TO CARRY FORWARD

Record the result at the strength the evidence supports, together with its source, scope, and important limits. Name the decision it can inform and who owns that decision. If another answer could change what happens next, state the uncertainty and the nearest credible check. If another team owns the next step, carry the bounded result to them and record their response. If no further investigation or information change is warranted now, say why and what would reopen the question. Use the Business Impact Readout when the result needs to travel or be reviewed.

Garrett French

Garrett French is the founder of Citation Labs, where he helps brands stay visible in AI answers and search through citation optimization and relevance-led link building at scale. His team studies how buyers use AI tools to shortlist purchases and deploy campaigns designed to increase client citations in recommendations.

He also built Xofu, a platform that tracks brand visibility across AI-generated recommendations, benchmarks competitors, and surfaces the pages AI references. And he leads ZipSprout, which builds sponsorship links by connecting businesses with nonprofits, events, and local organizations.

Garrett’s current explorations focus on decision efficiency and AI response behavior: how buyers decide, how AI systems “decide,” and how comparison assets influence what is cited for high-intent selection prompts.