The Metric You Can Measure Is Not Always the Result You Should Manage

A metric can be useful and still be the wrong basis for investment.

That distinction matters because organizations naturally manage what their systems can see.

Clicks can be measured. Leads can be counted. Appointments appear in a scheduling system. Response rates show up quickly. Activity can be assigned to teams and reviewed every week.

The business result often sits farther downstream.

The visible metric can stop too early

At ATI Physical Therapy, marketing systems could show which keywords produced clicks, which campaigns produced conversions, and which leads became appointments.

Those measures helped explain what was happening at the top of the funnel.

They stopped before the business outcome.

An appointment did not produce revenue by itself. A prospective patient had to become a patient and begin care.

The relevant path was keyword to click to appointment to patient to revenue.

Once that path became visible, the investment question changed.

Should marketing dollars go to the sources that produce the most appointments, or to the sources that produce the most valuable patient outcomes?

Those answers were not always the same.

Connecting marketing-source information to downstream patient and revenue data gave leaders a better basis for allocating spend. Revenue increased by 10% without an increase in marketing spend.

The important change was the decision, not the dashboard.

Proxy metrics still matter

I would not remove the intermediate measures.

A business needs to know whether demand is being created, whether prospects are responding, whether appointments are being scheduled, and where conversion changes.

Those measures help diagnose the system.

The problem begins when the proxy is treated as though it were the result.

A lead may indicate interest.

An appointment may indicate willingness to engage.

A response rate may show that a message is relevant.

An activity measure may show that a team is working the process.

Each can be valuable.

When leaders are deciding where to put money, people, or executive attention, I want to keep moving downstream and ask what business result that activity produces.

Different functions often manage different pieces of the same outcome

This gets harder in large organizations because no single team may own the full path.

One group owns media. Another owns CRM. Another owns sales or field execution. Another owns analytics. The final business outcome may appear in a completely different operational system.

Each function develops measures it can influence.

That is reasonable, but it can create local success without improving the overall result.

I saw that in Army recruiting.

Marketing could generate response. CRM could communicate with prospects. Recruiters could work leads. Analytics could report performance within individual areas.

The result leadership ultimately cared about was enlistment.

A prospect had to progress through several stages before that happened.

Looking across the complete journey changed the management question. Instead of asking only how to increase activity inside each function, we could ask what helped a qualified prospect move to the next stage.

The upstream measures remained useful. They were placed in the context of the outcome they were supposed to support.

The metric that governs allocation is the one that matters most

A dashboard may contain dozens of measures.

I pay close attention to the one that changes a decision.

What causes a leader to move budget?

What causes a team to change a process?

What causes an initiative to receive more resources?

What causes an executive to conclude that performance is improving?

If the highest-volume lead source gets more budget, the organization is managing lead volume.

If the channel with the lowest appointment cost gets more budget, it is managing appointment cost.

If those measures have a reliable relationship to the business result, that may be sensible.

If the relationship is weak or unknown, the organization may be optimizing what is easy to see.

Perfect measurement is rarely required

The answer is not to wait for a flawless attribution system.

Customer journeys cross systems. Data is incomplete. Some outcomes take months to appear. External factors affect results. The connection between activity and value is often imperfect.

I care more about whether the organization can make a better decision than it could before.

Can we connect activity to a more meaningful downstream result?

Can we distinguish sources that create volume from sources that create value?

Can we see where progression stops?

Can we identify which investment is producing the economic outcome the business needs?

At ATI, we needed enough connection between marketing activity and revenue to make better investment decisions. Perfect attribution was unnecessary.

That was enough to change where the money went.

Ask what the metric is standing in for

When a measure becomes prominent in an organization, I ask a simple question:

What result is this metric supposed to represent?

Then I keep moving downstream.

Eventually the chain reaches the outcome the business exists to produce: revenue, margin, retention, successful adoption, an enlistment, a completed service, or another result with economic or operational value.

The intermediate measures help explain how the system is performing.

The downstream result should govern the consequential decisions whenever the business can connect the two with reasonable confidence.

Measure the proxies.

Manage the outcome.