Measuring What Is Working

A campaign can show strong numbers and still be doing damage. Metrics that measure what happened without measuring how the customer felt about it are only telling half the story.

Watch Breakdown

Most businesses measure performance through outputs — clicks, conversions, cost per acquisition, revenue. When those numbers look good, the conclusion is that things are working. When they decline, the conclusion is that something needs to change.

 

But output metrics only reveal what happened. They do not reveal why it happened, whether it should be replicated, or whether the result was actually good for the business in ways that extend beyond the immediate transaction.

A campaign can produce high conversion while eroding trust. Revenue can grow while satisfaction quietly declines. A funnel can perform well by the numbers while creating exactly the kind of buyer experience that produces weak retention, low referrals, and the kind of reputation that limits future growth. And none of those problems appear in the output metrics until long after the behavior that created them has already been running for a significant period.

THE FUNDAMENTAL

 

  • Measuring what is working requires measuring more than what happened. It requires understanding how the customer experienced what happened — whether trust increased or decreased, whether expectations were met or created confusion, whether the relationship built through the transaction is the kind that produces loyalty and referrals or the kind that produces one-time transactions and silence.

    This is the principle that determines whether performance measurement is telling the business what it actually needs to know or only what is easiest to quantify — and the difference between those two things determines whether the decisions that follow from the measurement make the business stronger or optimize it in the wrong direction.

    When performance is measured through both outputs and customer-aligned signals — when trust, satisfaction, clarity, and perceived value are tracked alongside clicks, conversions, and revenue — the picture is complete enough to make decisions that genuinely improve the business. When only outputs are measured, decisions optimize for outputs that may be improving while the customer relationship they depend on is deteriorating.

  • High conversion can coexist with declining trust. Revenue growth can mask customer dissatisfaction. Strong engagement metrics can accompany weak perceived value. These are not theoretical possibilities — they are the consistent pattern of businesses that optimize for measurable outputs while leaving the customer's actual experience unmeasured and therefore unmanaged.

    The customer's perception of the business is what determines whether they return, whether they refer others, and whether the business grows through compounding trust or through constant acquisition of new buyers to replace those who did not find the experience worth repeating. And perception is shaped by every interaction the customer has — not just the conversion moment that the output metrics capture.

    Measuring only outputs and assuming that good numbers mean a good customer experience is like measuring a relationship by the number of conversations that happened rather than by whether those conversations made both parties feel understood. The number tells you something. It does not tell you whether what happened was actually good.

  • Most businesses measure what is easy to measure rather than what is most important to understand. Clicks, conversions, and revenue are easy to track because they are quantifiable and visible. Trust, satisfaction, and emotional response are harder to track because they require a different type of observation — but they are the signals that determine whether what the output metrics are reporting is actually building something durable or just capturing short-term transactions.

    Common mistakes include:

    Interpreting strong output metrics as evidence that everything is working without examining whether the customer experience behind those outputs is creating the kind of trust and satisfaction that sustains long-term growth.

    Optimizing campaigns for the metrics being tracked rather than for the outcomes that matter — which produces campaigns that improve their own scores while potentially degrading the customer experience they are creating.

    Collecting customer feedback through surveys and reviews but not integrating it into the performance evaluation that determines what changes — which means perception data exists without influencing the decisions that shape what customers experience.

    Separating marketing performance from customer experience as if they are distinct functions rather than recognizing that every marketing interaction shapes the customer's perception of the business in ways that affect every subsequent interaction.

    Noticing that trust or satisfaction is declining only after it appears in retention or referral metrics — which is the last indicator rather than the first, and by which point the behavior that produced the decline has already been running for a significant period.

    Metrics that measure outputs without measuring the customer experience behind them are useful for identifying what happened. They are not sufficient for understanding whether what happened was actually good for the business.

  • Performance measurement is complete when it answers two questions simultaneously — what happened and whether what happened was good for the customer relationship. Output metrics answer the first. Customer-aligned signals answer the second. Both are necessary for decisions that genuinely improve the business rather than optimizing it in directions that feel productive but erode what the business is actually trying to build.

    Customer-aligned signals are the metrics that reflect how the experience felt from the customer's perspective — whether they felt understood, whether expectations were met, whether they left the interaction with more confidence and trust than they arrived with, whether the value they received matched or exceeded what they were led to expect. These signals cannot be fully captured in click-through rates and conversion percentages. They require a different type of tracking that goes beyond what happened to understand why it happened and how it was experienced.

    When both types of measurement exist, they check each other. A campaign with strong output metrics and weak customer satisfaction signals is a campaign that is producing short-term results at the cost of long-term relationship quality. A funnel with lower conversion but stronger trust and satisfaction signals may be producing buyers who are more loyal, more likely to refer, and more likely to purchase again — and whose lifetime value makes the lower initial conversion rate the better outcome.

    The decisions that follow from complete measurement are different from the decisions that follow from partial measurement. And those different decisions, accumulated over time, produce significantly different businesses.

  • Campaigns continue past the point where they should have been adjusted because the output metrics that trigger review are still acceptable while the customer experience metrics that would have signaled problems are not being tracked. Trust declines without being detected until it appears in retention numbers — by which point the behavior that produced the decline has already shaped how a significant number of customers perceive the business.

    Strategy gets optimized in the direction of the metrics being tracked rather than in the direction of what would actually make the business better — and the metrics being tracked are the ones that were easy to measure rather than the ones that are most important to understand. Over time, decisions made on incomplete information consistently move the business slightly further from genuine strength and slightly closer to the kind of performance that looks good on a dashboard while the foundation it depends on quietly weakens.

 
Make it

VIDEO SECTION

Information

Embed Block
Add an embed URL or code.

APPLICATION / WHAT THIS LOOKS LIKE

 

A business runs a campaign that produces strong click-through rates and solid conversion. The team reviews the output metrics and concludes the campaign is performing well. It continues running without significant adjustment.

Six months later, the team notices that the clients who came in through that campaign have lower retention rates than clients from other sources. Referrals from that group are rare. The lifetime value of those clients is lower than the business's average despite the conversion cost being similar to other campaigns.

The campaign looked good by output metrics and was actually damaging long-term performance in ways those metrics never revealed. The conversion was real but the experience that produced it — the message that attracted those buyers, the expectations it set, the experience they had after converting — was creating the kind of relationship that did not sustain itself.

If the business had been tracking customer satisfaction and trust signals alongside the output metrics, those signals would have surfaced the problem significantly earlier — not at the point where it appeared in retention and lifetime value, but at the point where customers were first signaling through their behavior and feedback that the experience was not matching their expectations. The campaign could have been adjusted before the damage accumulated.

Now compare that to the same business with a hybrid measurement system in place. Output metrics are tracked as before. But alongside them, customer satisfaction after the first interaction is measured, perceived value at the point of conversion is assessed, and clarity of the experience is evaluated. When the campaign produces strong conversion but weak satisfaction scores, the signal appears immediately — before the retention problem has time to develop. The message is adjusted to set expectations more accurately. The experience is modified to deliver more clearly on what the conversion implied. The output metrics stay strong while the customer-aligned signals also improve.

The campaign was the same. The measurement was not. And the measurement determined whether the damage accumulated or was caught before it did.

WHAT THIS MAKES IMPOSSIBLE

When performance is measured through both outputs and customer-aligned signals, it becomes impossible for a campaign, funnel, or strategy to appear to be working while actually damaging the customer relationship — because the customer-aligned signals will reveal what the output metrics cannot.

It becomes impossible to optimize for the wrong outcomes when both types of metrics are being evaluated and checking each other. It becomes impossible to notice trust declining only after it appears in retention numbers when satisfaction and perception signals are being tracked continuously rather than only inferred from outcomes that arrive later. And it becomes impossible to make decisions confidently on the basis of partial information when the complete picture — both what happened and how customers experienced it — is consistently available.

Measurement that captures only outputs is not wrong. It is incomplete. And decisions made on incomplete information are decisions made with a blind spot that will eventually produce the consequences that the missing information would have prevented.

COMMON MISTAKES

 

Most businesses weaken their strategic decision making by measuring what is easy to quantify rather than what is most important to understand — and by treating strong output metrics as sufficient evidence that the customer experience is also working.

Common mistakes include:

Assuming that high conversion means the customer experience is positive — which ignores that conversion measures a transaction moment while trust and satisfaction measure the relationship that determines whether the transaction leads to anything beyond itself.

Optimizing campaigns for the metrics being tracked rather than for the underlying outcomes those metrics are supposed to represent — which produces performance that improves by its own measures while potentially degrading the customer relationship it depends on.

Collecting customer feedback in isolation from performance review rather than integrating it as a data source that informs the same decisions that output metrics inform — which means perception data exists without influencing the strategy it should be shaping.

Tracking metrics without interpreting what they mean in terms of the customer experience — which produces reporting without understanding and decisions that address the metric rather than the reality behind it.

Waiting for retention, referral, or lifetime value signals to reveal customer experience problems — which are the last indicators rather than the first, and which arrive after the behavior that produced them has already run for a significant period.

Performance that looks good by its own measures and is actually degrading the relationship that sustains long-term growth is not good performance. It is measurement that is optimizing in the wrong direction because the most important signals were not included in what was being measured.

HOW TO KNOW IT’S WORKING

 

Performance measurement is complete when it reveals both what happened and whether what happened was good for the customer relationship — when the business can confidently say not just that conversion increased but that trust, satisfaction, and perceived value moved in the same direction.

Test it against five questions:

Are customer trust and satisfaction being measured alongside output metrics? If the performance review only includes what happened without including how customers experienced what happened, the measurement is partial and decisions made from it are missing the most consequential signals about whether the business is building something durable.

Do the metrics being tracked reflect emotional impact or only behavioral output? If every metric in the review can be answered by a number without any of them requiring an understanding of what the customer felt, the measurement system does not include the signals that determine whether what is being optimized is actually good for the customer relationship.

Are insights from customer perception being translated into specific strategy adjustments? If satisfaction scores, trust signals, and perception feedback are collected but do not consistently produce changes in messaging, funnel structure, or campaign strategy, the measurement is informing understanding without influencing decisions — which means the loop is open rather than closed.

Would the business notice if trust was declining before it appeared in retention metrics? If the honest answer is no — if the first signal of trust decline would be lower retention numbers rather than the behavioral and perception signals that precede them — the measurement system is reacting to outcomes rather than detecting causes while they are still correctable.

Is performance trending in both output and customer experience dimensions simultaneously? If output metrics are improving while customer satisfaction signals are declining, the business is optimizing for transactions while degrading the relationship. If both are improving together, measurement is complete and the improvement is genuinely building the business rather than just improving its numbers.

If performance is being measured through both what happened and how customers experienced it, decisions are being made with the complete picture rather than with a partial one. If output metrics are the primary or only input into performance review, the measurement is telling the business what it can most easily measure rather than what it most needs to know.

Browse