Marketing effectiveness is only valuable when it changes what people do

The results are clear; the team agrees with the analysis; the slides explain the brand story, the channels that worked and the likely upside of optimising. Then the deck is emailed around, and nothing changes.

That's not a measurement problem, it is an action problem.

I've seen this plenty of times with marketing effectiveness and MMM projects. The work is good, findings well received and everyone leaves the room agreeing that the recommendations make sense.

Six months later, the same media plan is running, the same promotional decisions are being made and the model is being treated as an interesting one-off rather than a way of running the business.

The gap is rarely caused by poor analysis: it's caused by the fact that nobody has turned the insight into a practical next step, agreed who owns it or worked out whether it is worth doing before the next urgent thing takes over.

“Spend more on Social” is not a next step

Imagine the model says annual Social spend should increase from £3m to £5m.

That may be a sound recommendation, but it's also a large commitment. A CMO may agree with the analysis while still worrying about the risk, finance may want evidence that the return will hold at the higher level and the media team may already have commitments in place.

Putting “increase social spend” on a slide does not resolve any of that.

The next best action could be to upweight Social in two regions by £50k, agree what success looks like before the activity runs and assess whether the additional spend delivers the expected result.

That is a much more deployable recommendation.

It gives the business a controlled way to build confidence before committing another £2m. If the result holds, the larger decision becomes easier. If it does not, you have learned something important without making a costly mistake.

This is the point where marketing effectiveness needs to become more commercial. The job is not simply to explain what the model says, it is to help the business decide what it can sensibly do next.

A long list of recommendations is usually where action stops

Most projects produce more than one answer. There may be a case for increasing brand activity, changing the promotional calendar, reducing spend in a channel with falling returns or moving budget between products, regions or audiences.

All useful, maybe, but a list of recommendations is not a priority plan.

When every action is presented with the same status, the team has to make an informal judgement about what is worth doing. That judgement usually happens under pressure, alongside sales targets, agency commitments, budget cycles and the day-to-day work that already fills everyone’s diary.

The least disruptive action often wins, not the most valuable one.

The answer is to rank actions properly, using three things:

likely profit upside effort required across marketing, finance, sales and agencies confidence in the result, including whether a test is needed first

• likely profit upside

• effort required across marketing, finance, sales and agencies

• confidence in the result, including whether a test is needed first

The analytics should show the likely value, the people who run the business need to bring the practical reality.

An analyst can see an opportunity in the numbers, but they may not know that it depends on retailer agreement, production capacity, legal approval or a media plan already signed off. Equally, a commercial team may know an action is difficult but not realise the amount of profit being left on the table by doing nothing.

You need both. High-value, low-effort actions should move first:

High-value actions with less certainty may need a test. Low-value actions that require a huge amount of work should stop taking up space.

• High-value actions with less certainty may need a test.

• Low-value actions that require a huge amount of work should stop taking up space.

That is not complicated, but it is often the missing link between an insight and a decision.

Recommended by LinkedIn

The model needs an owner, a deadline and a place to live

The most important output from a marketing effectiveness project should not be a deck. It should be a small number of agreed actions, each with a named owner, a date for the first step, an expected value and a way of seeing whether it has happened.

If nobody owns the action it will drift.

If it has no deadline, it will sit behind work that feels more urgent.

If it lives only in PowerPoint (or google slides!), it will disappear when the meeting ends.

Actions should show up in the tools people already use to manage work, whether that is Monday, Asana or something else. They need to sit alongside the other commercial priorities, rather than relying on someone remembering what the model said three months ago.

This also gives leadership a much better view of the value being created.

A project may identify a meaningful profit opportunity, but that is only theoretical until decisions are made and activity is put into market. Tracking deployed actions against expected value makes the difference visible.

It also creates useful accountability - you can see whether the issue is the recommendation was wrong, the action was not taken or the result did not materialise as expected - three very different conversations.

Trust comes from taking sensible risks, not explaining more charts

There is a tendency in analytics to assume that trust is earned through technical detail.  More charts, more explanation and more complex modelling language rarely solve the problem, they can make it worse.

Trust comes when people understand the commercial decision in front of them, can see the likely upside, understand the risk and have a sensible way to test it.  That is why a smaller regional test can be more valuable than a large recommendation that never gets deployed.

The test brings marketing, finance and the wider business into the work. It turns a modelled result into something the team can judge in its own market, using its own sales and profit numbers.

It also shifts the conversation from “do we believe the model?” to “what have we seen happen when we acted on it?”  Which is where models start to earn their place in planning and budgeting.

Build the project around the decisions you need to make

Before starting an MMM or marketing effectiveness project, I would ask a few questions early.

What decisions do we need to make when this work is complete? Who will make them? What could realistically change in the next quarter? Which recommendations would need testing before a bigger commitment? How will we know whether the action happened, and whether it worked?

• What decisions do we need to make when this work is complete?

• Who will make them?

• What could realistically change in the next quarter?

• Which recommendations would need testing before a bigger commitment?

• How will we know whether the action happened, and whether it worked?

These questions shape the work from the start.

They influence the data you collect, the teams you involve and the outputs you build. They also stop measurement being designed around an agency review meeting and then start designing it around how the business actually makes decisions.

The aim is not more recommendations.

It is fewer, better actions that get made, tested and tracked.

Summary

Marketing effectiveness should not end with a deck that makes sense. It should end with a set of actions, owned by real people, with dates, expected value and a route into the team’s normal way of working.

If your latest marketing effectiveness results are sitting in an email attachment, ask one simple question: what will somebody do differently because of this next week?

If there is no clear answer, the work is not finished.