The last mile of measurement

One marketing leader described receiving findings in a PowerPoint attachment that ended up “saved into my folder somewhere”.

Their verdict was simple: “It’s not made it easy to take action.”

Marketing effectiveness projects can contain excellent analysis, careful modelling and pages of useful findings, then still fail to change a single decision.

The problem is rarely that a team needs more numbers., it’s more about getting useful evidence to the right people, at the right time, in a form they can trust and act on.

That is the last mile of measurement - moving from a recommendation to something that is deployed and then measured again. That should be the real test of any marketing effectiveness project.

A model can be right and still be too far from the business

An infrequent model update can be useful for annual planning, but it cannot answer every question that arrives when budgets, forecasts or performance shift.

A deck also does not become useful because it has been put into a dashboard.

People still need to understand what changed, why it changed and what they can do about it.

One measurement leader put the challenge well, . they stated “Tthe issue is not simply access to the work, it’s the marketing knowledge, commercial judgement and ability to tell a compelling story with it.”

That is why a model that lives with one analyst, agency or consultant struggles to change behaviour. Everyone else waits for an explanation, the moment passes and the team returns to the familiar report.

The project has produced output, but it has not created action.

Short-term results can crowd out the decisions that matter

Performance media has a built-in advantage - its results appear quickly, it sits close to the transaction and it produces numbers people can read every day.

Brand investment works differently.

Its effect may appear through future demand, consideration, price sensitivity and helping the return from performance activity. Those effects can be harder to connect to a decision happening this week, so they are often treated as less certain and cut first.

A global brand leader described strong measurement at the top and bottom of the funnel, with “almost no measurements in between”.

That creates a real commercial problem because.  Iif brand is building the pipeline of future demand, how do you show that in the metrics used today?

The answer is not a generic long-term multiplier.

It is to put brand and performance into the same commercial view, showing both immediate and longer-term return in revenue and profit terms. Finance can then see the trade-off rather than being asked to take brand on faith.

More reports do not create more trust

MMM, attribution, lift testing and last click reporting all answer different questions.

MMM gives a broad view of incremental impact across channels and can show persistent effects Lift tests help verify disputed results or test smaller spend levels with lower risk. Last click gives an immediate, detailed operating read.

• MMM gives a broad view of incremental impact across channels and can show persistent effects

• Lift tests help verify disputed results or test smaller spend levels with lower risk.

• Last click gives an immediate, detailed operating read.

The problem begins when each sits in a separate report with a separate number.

One leader described the search for a single source of truth as finding methods that are “less wrong, like, depending on what your question is.”

That is refreshingly honest.

There is no value in pretending every technique should produce the same answer. There is value in making their roles clear, calibrating the evidence and bringing it back to one commercial view.

Recommended by LinkedIn

Without that, the CEO sees competing answers and defaults to the one they already know, usually last click.

Marketing then has a more sophisticated answer that nobody is prepared to use.

A media answer alone is rarely a business answer

A change in marketing spend may affect sales, but price, promotions, distribution, retention, new products and margin determine what that change is worth to the business.

So the useful question is not simply, “What was the ROI?”, it'sits “What happens to the plan and the P&L if we do something different?”

How does an extra pound in marketing compare with a promotion, a price change or a new product launch?

What does it do now, and what does it do over time?

Showing marketing in that context matters because it changes the conversation. Marketing stops presenting a channel result, finance and commercial teams can see the full trade-off, then the board has a decision it can use.

That is how marketing effectiveness earns a place beyond the marketing function.

Actionability starts before the model

The most useful shift is simple: start with the decision.

Before the work begins, agree:

Which commercial decision needs to change Who will own that decision When they need the evidence What action they could realistically take How the result will be tracked in revenue, profit or another agreed commercial measure

• Which commercial decision needs to change

• Who will own that decision

• When they need the evidence

• What action they could realistically take

• How the result will be tracked in revenue, profit or another agreed commercial measure

The model needs to sit close to the business, using commercial metrics and involving finance, sales and marketing rather than living in one function. It needs to measure the things that shape the decision, not only the channels that are easiest to report.  It needs to have non-conflicting numbers - a measurement system that uses each techniques strengths to deliverget to  realistic, holistic and timely results.

It also needs to leave the project with a ranked set of actions.

Each action should have an expected commercial value, a realistic level of effort and a named owner.

Then the action needs to go into ClickUp, Asana, Monday or the tool the business already uses, assigned against individual names and held to account.

Once deployed, the action needs to be tracked as to whether it happened or not, and therefore what commercial upside has been observed vs the plan.

Otherwise, the final slide becomes the end of the work - it should be the start.

Measurement work should generally find potential profit gains of £1m to £20m per project, but that value only reaches the P&L when people make and follow through on better decisions.

Summary

The quality of a marketing effectiveness project is not defined by the model, the dashboard or the sise of the deck.

It is defined by whether the business can use the evidence to make a better decision, assign the action and see whether it worked.

That is the last mile of measurement - it doesn’t end with a deck, it ends when the action is deployed and the uplift from the insight is realised.