Measurement That Works: A Guide for Challenger Brands

Introduction: The 5% problem

Here’s a question we like to ask brand leaders. Your sales are up 5% this year. Are you doing well?

Most people say yes. But what if your competitors are up 20%? Then the category is growing four times faster than you are, and you’re quietly losing share. You’d be celebrating a year in which you fell behind, and nothing in your own numbers would tell you.

That’s the 5% problem, and it’s a measurement problem. Not a marketing one.

Most brands don’t lack data. They’re drowning in it. What they lack is a clear, timely answer to three questions:

1. Are we gaining or losing momentum against the brands we compete with?
2. What’s driving it?
3. What should we do about it, while there’s still time to act?

Almost nothing in the typical measurement stack answers those questions directly. This guide is about how to get there.

Who this guide is for

We wrote it for the people running challenger brands. Marketing leaders who have to defend every dollar to a CFO. Founders who are, more often than not, also the acting CMO. Teams of five or fifteen, not fifty, with no data science function and no econometricians on call.

Enterprise brands can throw people at broken measurement. Challengers can’t. Less room for a wrong call, less time to wait for an answer, and the same hard questions from the board.

What’s inside

We start with why marketing measurement is so dysfunctional. Then we set out five principles for measurement that actually works for a challenger brand:

• Behavioral. Built on what people do, not what they tell a survey.
• Long and short together. Brand building and activation on one scale.
• Competitive. A like-for-like view of the brands fighting for the same buyer.
• In the CFO’s language. Connected to revenue.
• Always-on and actionable. There when the decision is, with an answer rather than another chart.

Next, we look at the newest blind spot in marketing: the customer who’s disappearing behind AI assistants. Then we bring the principles together where most teams first feel the gap: setting campaign KPIs. And we finish with a 30-day audit you can run on your own measurement, with a scorecard to see where you stand.

None of this requires buying anything. Most of it starts with asking better questions of what you already have.

You don’t win a category on your own scoreboard.

Chapter 1: Why marketing measurement is broken

Marketing measurement is dysfunctional. It’s fragmented across a dozen tools, disconnected from revenue, and trusted by almost nobody who signs off on the budget.

More than a century ago, John Wanamaker is supposed to have said that half his advertising money was wasted, and the trouble was he didn’t know which half. Brands now spend fortunes on tools built to answer that question. Most still can’t.

Rory Sutherland put his finger on why: we’ve become richer and richer on data, and poorer and poorer on meaning.

A wall of tools that don’t talk to each other

In 2023, Gartner found marketers were using just 33% of the capabilities in their martech stack, down from 58% in 2020. That isn’t laziness. The stack was never built to fit together.

A typical challenger brand pays for performance reporting, attribution, a brand tracker, social listening, web analytics, NPS, the odd brand lift study and, if it can afford one, a marketing mix model. Then it builds a monthly deck to report on the reports.

Each tool is good at something. The trouble is they answer different questions, in different currencies, on different clocks.

None of them was designed to talk to the others. None of them, on its own, tells you what’s moving sales and why.

Nobody owns the join

Attribution never agrees with the platforms. Awareness goes up while sales stay flat, and nobody can say why. A dozen sources get squeezed onto one slide for the CEO.

At most challenger brands, that reconciling lands on a marketing team with no data science function. They end up doing analytics in the hours left over from doing marketing.

The output describes what happened and says almost nothing about what to do next. And because every tool tells a slightly different story, the CFO trusts none of them. You pay for the data and still decide on instinct.

What it costs: budget, influence and, eventually, the job

Gartner calls it the brand doom loop. Companies skimp on brand measurement, stop trusting what it tells them, and so win even less funding the next time around. Gartner found 84% of companies are caught in it, and they’re half as likely to exceed their growth targets as companies that can value their brand.

The pressure lands on the CMO. Average CMO tenure at S&P 500 companies is 4.1 years, against five for the C-suite as a whole, according to Spencer Stuart. At consumer companies, it’s 3.5 years, the shortest of any sector. Measurement isn’t the only reason. But a marketing leader who can’t show what marketing returns walks into every budget meeting on the back foot.

Why challenger brands pay the most

Because they can’t paper over it.

Enterprise brands throw people at the problem: analytics teams to stitch the sources together, econometricians on call, budget to run every study twice. Even they struggle. In Gartner’s 2026 CMO Spend Survey, most respondents worked at companies with more than $1 billion in revenue, and 56% still said they lacked the budget to deliver their strategy.

A challenger has none of that cover. Challenger brands rarely lose because their marketing is worse. They lose because they can’t see clearly enough, quickly enough.

More data isn’t the answer. Neither is certainty.

Most brands are already swimming in data. The problem is getting from the signals you already generate to what’s happening, what it means and what to do next, while the answer still matters.

Certainty isn’t the answer either. Les Binet has called the promise of certainty in measurement snake oil. Marketing is probabilistic, and anyone promising certainty is selling something.

What’s missing is a set of principles for what good measurement looks like. That’s what the next five chapters are for.

If you’re a founder: you probably don’t need more tools. You need fewer, that answer the questions you actually ask on Monday morning.

If you’re a CMO: the question isn’t whether your stack is good. It’s whether anything in it answers the CFO’s question: what is our marketing doing for revenue?

Chapter 2: Principle 1 — Watch what people do, not what they say

There are two ways to find out how your brand is doing. You can ask people, or you can watch what they do. Most brand measurement asks.

Trackers, brand lift studies and NPS all run on surveys. They’re useful color. But they measure attitude, usually once a quarter and usually with a small sample. And stated intentions are a notoriously poor guide to what people actually buy.

The zero-party promise

For the last few years, the industry’s favorite answer has been zero-party data. It’s Forrester’s term for information a customer intentionally shares with a brand: quiz answers, preference centers, onboarding questions, loyalty profiles.

It has real strengths. It’s built on consent, it’s explicit, and it’s yours. For personalization and CRM, it’s genuinely valuable.

But as a way of judging whether your brand is winning, it has three problems.

1. It only reaches people who’ve already found you. It needs a touchpoint: a site visit, an app, an email sign-up. The buyer who chose a competitor never got as far as your quiz.
2. It’s what people say, not what they do. Declared preferences are attitude, just collected more politely.
3. It’s skewed toward your most loyal customers and blind to competitors. The people who fill in your preference center are your biggest fans. Your competitors’ customers never will.

Put bluntly, zero-party data is a survey with better branding.

First-party data is a mirror

First-party data has the same limit. It shows you the customers you already have.

As Byron Sharp’s work at the Ehrenberg-Bass Institute shows, brands grow by winning more buyers, mostly light buyers who barely think about you. They’re precisely the people least likely to be sitting in your CRM.

So the data you own tells you a lot about the customers you’ve already won. It tells you very little about the ones you need to win next.

What behavior can see that surveys can’t

Even buyers who never visit your site leave traces in aggregate: what scans at the till, what people search for, which sites they visit, what gets said about brands in your category, and how often AI assistants recommend each brand.

Measured across a whole category, that trail is the one view where every brand shows up like-for-like. Nobody has to opt in for their purchase to scan.

Keep each kind of data in its lane

None of this means throwing your surveys or your CRM away. It means using each for what it’s good at.

• Zero- and first-party data: the relationship. Personalization, service, loyalty.
• Surveys and trackers: the why behind perception. Useful context, especially for creative and positioning.
• Market-level behavior: the scoreboard. Whether your brand is actually winning, against everyone else, in what people do.

What to do this quarter

1. Sort your metrics into “ask” and “observe.” Most teams find their brand evidence is almost entirely asked.
2. Find one behavioral signal for brand health. Share of search is the easiest place to start, and it’s free.
3. Stop using loyalty data to judge growth. It describes the customers you have, not the ones you need.

Opinions are useful color. Behavior is the scoreboard.

Chapter 3: Principle 2 — Measure the long and the short together

Les Binet and Peter Field made a case for the IPA that most marketers now take as read: brands need both. Short-term activation that sells this month, and long-term brand building that makes next year’s sales easier. Cut either one and you pay for it eventually.

Very few people argue with that anymore. The trouble starts on Monday morning, when you try to measure it.

Two scoreboards, two clocks

In most marketing teams, the long and the short live in different worlds.

The short term belongs to performance tools: clicks, conversions, attribution, incrementality tests. They observe real behavior, which is good. But they mostly see the last week or two, and mostly your own channels.

The long term belongs to brand tools: trackers, lift studies, NPS. They ask people what they think, usually quarterly. Useful, but it arrives late, and it measures attitude rather than behavior.

So the marketing leader ends up with two scoreboards on two clocks. One is fast, behavioral and short-sighted. The other is slow, attitudinal and hard to connect to a sales number.

Guess which one the CFO trusts.

Why brand gets cut first

Budget season arrives. The performance numbers come with a revenue figure attached. The brand numbers come with an awareness score from last quarter.

It isn’t that anyone thinks brand doesn’t matter. It’s that brand’s evidence arrives late and in a different currency. When the conversation turns to “show me the money,” the line item that can’t answer gets questioned first.

That’s how challenger brands end up over-investing in activation. It works, visibly, right up until the pool of people who already know you starts to run dry. Kantar’s growth research found that neglecting exposure, reaching the buyers of the future, is the most punishing mistake a brand can make. The damage just takes a while to show, which is exactly what makes it so easy to ignore.

And the long tail is real. The IPA’s Profit Ability research found advertising effects can persist from four months to more than two years after a campaign ends.

Measure only one, and it will mislead you

This isn’t a plea for more brand budget. It’s a case for measuring both, because each one alone will steer you wrong.

• Measure only the short term, and activation always looks like the better bet. Keep following that chart and you end up harvesting a brand rather than building one, spending more and more to convert a shrinking pool of people who already know you.
• Measure only the long term, and you can see whether the brand is healthy, but you can’t tie it to sales. You tend to find out too late to change course. That’s no position to defend a budget from.

Whatever you measure, you’ll end up optimizing for. Measure only one, and you’ll optimize for only one.

The thread that connects them

The long and the short aren’t really separate. A useful way to see it is three clusters of signals, each on a different time horizon:

• Visibility: the signals that put you in front of future buyers. Reach, share of voice, search interest.
• Engagement: the signals around the moment of purchase. Site visits, consideration, conversion.
• Reputation: the signals that keep customers coming back and talking about you. Reviews, earned conversation, repeat purchase.

Visibility today tends to feed the engagement that converts next month, and the reputation that brings people back next year. Short-term signals are often the earliest reading you’ll get of long-term health. Split them across different tools on different clocks, and you lose the thread.

The empty corner

Plot the measurement landscape on two lines: whether a tool asks people or observes behavior, and whether it reads the short term or the long term. Brand tools sit in the “ask” half. Performance tools observe, but only the short term. The corner that matters most to a growing brand, long-term brand health measured from behavior, is the one most stacks leave empty.

Marketing mix modeling sits across the line: it observes sales, but mostly reads the short-term response, months after the fact.

What to do this quarter

1. Map your tools on the two lines. Ask or observe, short or long. See which corner is empty.
2. Put brand and performance on one slide, in the same currency. If they can’t sit side by side, that’s the gap your CFO feels too.
3. Track at least one leading indicator of visibility every month, not every quarter.
4. Agree the evidence before the budget review. Ask finance what would make brand spend defensible, then go and get it.

Binet and Field showed us we need both. The job now is being able to see both, at the same time, while there’s still time to act.

Chapter 4: Principle 3 — Measure against your competitors

Back to the 5% problem. Sales up 5% looks like a good year, until you learn the category grew 20%. Almost every brand measurement tool is built to look at one brand: yours. And the comparison that actually predicts growth is the one most teams skip.

The numbers that predict growth are all relative

The most robust findings in marketing effectiveness have one thing in common. They’re about your brand compared with everyone else’s.

Share of voice. John Philip Jones set out the pattern in the Harvard Business Review in 1990, and Les Binet and Peter Field later tested it against the IPA’s effectiveness data. Brands whose share of voice runs ahead of their share of market tend to grow. On average, ten points of excess share of voice is worth about half a point of market share a year. You can’t calculate that without knowing what your competitors spend.

Share of search. When Binet introduced it at the IPA’s EffWorks conference in 2020, his finding was that share of search can predict market share, sometimes up to a year ahead. When James Hankins later looked at 30 cases across 12 categories and seven countries for the IPA, share of search tracked about 83% of a brand’s share of market on average. It’s built on the competitive set by definition: your searches as a share of everyone’s in the category.

Your growth lives in your competitors’ customers

Byron Sharp’s work at the Ehrenberg-Bass Institute gives a deeper reason. Brands in a category share their buyers with each other, broadly in line with their size. Your customers also buy your competitors. Their customers also buy you.

So growth, for a challenger especially, mostly comes from winning a little more of the people who currently buy someone else. If you’re not watching those brands, you’re not watching where your growth will come from.

Why most measurement looks inward

It isn’t laziness. It’s how the tools were built.

A brand tracker surveys people about your brand, and maybe a handful of rivals, once a quarter. A marketing mix model looks at your own spend against your own sales, with competitors as a control variable at best. Dashboards show the data you own. Your competitors will never hand you their media plan.

But they do show up in the same behavioral data you do: in search, in conversation, in what scans at the till, and in what AI assistants recommend.

What a competitive view tells you

When you can watch your whole competitive set on the same lens, a different set of questions opens up.

• Who’s doubling down, and who’s pulling back? A rival going quiet is an opening. The same spend suddenly buys you a bigger share of voice.
• Where is the white space? A channel, a moment or a buying occasion nobody in the category owns.
• Is your gain real? Did you grow, or did the category grow and you just kept pace?
• Who’s coming up behind you? A small brand gaining momentum shows up in search and conversation long before it shows up in a share report.
• Are you loud, or effective? Two brands can be equally visible and get very different results. Knowing which one you are is worth more than another month of impressions.

What to do this quarter

1. Define your competitive set the way your buyer would. Not the brands you’d like to be compared with, but the ones on the shelf next to you and in the same search results.
2. Measure them the same way you measure yourself. Same signals, same cadence, same lens. A like-for-like view is the only one that holds up.
3. Track share, not absolutes. Share of voice, share of search, share of conversation and, increasingly, share of AI recommendations.
4. Bring it to every big decision. Budget, launch timing, channel mix. The question is never just “is this working?” It’s “is this working better than what they’re doing?”

If you’re a founder: start with share of search. It’s free, it’s behavioral, and it’s competitive by design. Pick five rivals and track it monthly.

You don’t win a category on your own scoreboard.

Chapter 5: Principle 4 — Speak the CFO’s language

Every marketing leader eventually sits across from a CFO who asks some version of the same question: what is our marketing doing for revenue?

If nothing you own can answer that, against your competitors, in time to act, you’ve found the real problem. Not the creative. Not the media plan. The evidence.

Different currencies, one budget sheet

The budget sheet is written in dollars. Performance marketing reports in dollars, or something close to them. Brand reports in awareness, consideration and sentiment.

None of those are wrong. They’re just not the currency the decision is made in. When brand evidence has to be translated on the fly in a budget meeting, it loses.

Les Binet’s advice is blunt: marketers need to start talking the language of finance. That doesn’t mean pretending brand is a spreadsheet. It means bringing brand evidence the CFO can read, in the currency the CFO already uses.

The stakes are higher than they look

Gartner predicts that by 2027, over 40% of CMOs who push for bigger brand budgets will lose influence in the C-suite, because they won’t be able to show the returns.

Read that carefully. The risk isn’t asking for brand budget. It’s asking without evidence. The marketing leaders who keep their seat at the table are the ones who can say, in plain numbers, what happens to sales if brand investment holds, grows or gets cut.

What finance actually wants to see

In our conversations with finance leaders, the asks are surprisingly consistent:

• A link to sales. Not proof of every dollar, but a credible connection between brand activity and revenue.
• Direction early. A signal that something is slipping while there’s still time to fix it, not a post-mortem.
• Context. How you’re doing relative to the market, so a good year isn’t just a rising tide.
• Honesty about uncertainty. Ranges and confidence, not false precision.

That last one matters. Marketing is probabilistic. A finance team will trust “we’re 80% confident this is directionally right” far more than a suspiciously exact ROI figure.

Beware the multiplier

One common habit deserves a warning. Many models capture only the short-term sales response, then estimate the long-term effect by multiplying it. A WARC paper on econometrics noted that many practitioners simply multiply the short-term return by four.

In other words, the most important number in your brand case is often an assumption someone typed into a slide. A sharp CFO will find it. Better to bring evidence that measures the long-term effect directly, even if it comes with a wider range.

What to do this quarter

1. Ask finance first. Before the next budget cycle, ask your CFO what evidence would make brand spend defensible. Write it down.
2. Translate one brand metric into revenue terms. Even a rough, honest estimate beats an awareness score.
3. Show it against the market. Growth relative to competitors is the context finance always asks for.
4. Retire any number you can’t explain. If you can’t say where a figure came from, don’t put it in front of finance.

If you’re a CMO: go into budget season with one slide that puts brand and performance in the same currency, against the same competitors. It changes the conversation.

Brand evidence the CFO can read is brand budget the CFO can approve.

Chapter 6: Principle 5 — Always-on and actionable

The questions we hear from challenger brand leaders are rarely annual ones. They’re weekly. A competitor launched on TikTok three weeks ago and is already showing up on your retail shelf. Search interest dipped last month. Should you hold spend, or cut it?

Measurement that answers those questions a quarter late isn’t measurement. It’s history.

The comeback of marketing mix modeling

Marketing mix modeling (MMM) is the clearest example of the timing problem, and it’s worth understanding because it’s back in fashion.

Thirty-odd years ago, barcode scanners let big consumer goods companies line weekly sales up against weekly spend for the first time. Then digital arrived, every click could be tracked, and MMM started to look like a slow, expensive relic. Now privacy rules, Apple’s tracking changes and the long goodbye to the cookie have broken user-level attribution, and MMM, which never needed to follow individuals, has walked back into the room. In an EMARKETER and TransUnion survey, 46.9% of US marketers said they planned to invest more in it.

MMM didn’t come back because it got better. It came back because the thing that replaced it broke.

What MMM does well

To be fair, MMM does something nothing else does. It estimates incrementality: the sales your marketing actually caused, net of price, promotions and seasonality. It models diminishing returns, so it can tell you where the next dollar stops paying back. Finance trusts it, and it’s privacy-safe by design. The new open-source Bayesian tools are a genuine step forward.

Where it falls short for a challenger

• It looks backward. A typical model uses two or three years of history and gets refreshed once or twice a year. By the time it tells you what worked, the moment to act has usually gone.
• It’s built for the short term. Most models catch the sales response within weeks of spend, and the long tail gets estimated rather than measured (see Chapter 5).
• It can’t see your competitors, beyond a control variable.
• It only models what you paid for. Earned conversation, search demand, a creator who picked you up unprompted, how often an AI assistant recommends you: hard to put in a spend column, so they tend to get left out of the model. They don’t get left out of the market.
• It needs scale. Open-source tools made the math free. They didn’t make the team free. With lumpy, concentrated spend, the model has less to learn from, and the uncertainty shows.

Verification layer, not monitoring layer

The most useful way to think about MMM comes from the econometrician Grace Kite, who argues it’s one key piece of the evidence, not the whole of it. She’s also candid about its weak spot: a lot of econometrics gets done and then not used. Only 28% of marketers say their organization is very effective at turning MMM insights into action, according to eMarketer.

So think of your measurement in two layers:

If you already run an MMM, keep it for the annual budget split. Just don’t expect it to answer this week’s questions.

Actionable means an answer, not a chart

Always-on isn’t enough on its own. A dashboard refreshed daily that nobody can interpret is just a faster way to be confused.

Good monitoring answers in the same shape every time: what’s happening, what it means, and what to do next. If a report can’t name a decision it changed, it’s decoration.

What to do this quarter

1. Write down the five questions your team asks most often. Next to each, note how long it takes to get an answer today.
2. Separate verification from monitoring. Make sure you have something in each layer, and that you’re not asking one to do the other’s job.
3. Apply the decision test. For every recurring report, name one decision it changed last quarter. No decision, no budget.

MMM is a very good answer to last year’s question. Challengers need an answer to this week’s.

Chapter 7: The new blind spot — AI visibility

For twenty years, the big promise of digital marketing was a direct relationship with your customer. Know who they are, see what they do, talk to them one to one. Whole direct-to-consumer categories were built on it.

That relationship is quietly disappearing. And what it’s disappearing behind is an AI assistant.

The customer you can’t see

More and more buying journeys now start, and sometimes nearly finish, inside a conversation you’re not part of. Someone asks ChatGPT, Gemini or Perplexity which sunscreen to buy. The assistant compares, shortlists and explains. By the time the customer reaches your website, if they reach it at all, the decision is mostly made.

You can see the shift even in ordinary search. SparkToro’s 2026 research, built on Similarweb clickstream data, found that 68% of US Google searches in the first four months of the year ended without a single click. When an AI Overview appears, click-through rates drop by nearly 60%. Inside a chat assistant, there often isn’t a results page at all.

No pixel fires inside that conversation. No cookie, no session, no email capture. The question they asked, the brands they compared and the reason you didn’t make the shortlist all happen out of sight.

Every brand becomes a wholesaler

Retail brands know this feeling well. The customer at the Target or Ulta checkout has always belonged to the retailer. AI is extending that to everyone, with someone else owning the moment of choice.

Brands have pushed back. When OpenAI launched in-chat checkout in 2025, it was scaled back within months, with purchases sent back to merchants’ own sites. Brands kept the sale. But the comparison, the shortlist and the moment of choice still happen inside the assistant. You can keep the sale and still lose sight of the customer.

Attribution was already struggling

None of this is happening to a healthy measurement system. Since 2021, iPhone apps have had to ask before tracking people across other apps, and a large share of people say no.

Add journeys that now happen inside AI conversations, and user-level attribution ends up measuring a shrinking, unrepresentative slice of buyers: the ones who still happen to be trackable. That isn’t measurement. It’s looking where the light happens to be.

What you can still see

Here’s the good news. You may not be able to see the individual customer, but you can still see the market.

As Chapter 2 showed, buyers who are invisible to you still leave traces in aggregate. And AI assistants themselves can be measured: how often each brand in your category gets recommended, for which questions, and in what company. That’s a new signal, and it belongs alongside share of voice and share of search.

What to do this quarter

1. Ask the assistants yourself. Run the ten questions a buyer in your category would ask across ChatGPT, Gemini, Claude and Perplexity. Note who gets recommended, and how often you do.
2. Put share of AI recommendations on the scorecard. Track it monthly, against the same competitive set as everything else.
3. Accept you won’t see every customer. Rebuilding individual tracking in a world designed to prevent it is an expensive way to measure a shrinking minority.
4. Measure the market for decisions. Use your own data for the relationship, and market-level data to judge whether you’re winning.

If an assistant shortlists three brands in your category and you’re not one of them, nothing in your own data will ever tell you.

You may never see every customer again. You can still see the market. Measure that.

Chapter 8: Benchmarking — Set campaign KPIs against the market

A campaign can hit every KPI on the wrap report and still lose. Reach, engagement and cost per click all come back green, while the biggest competitor in the category grows faster and takes share. The KPIs weren’t wrong because the team missed them. They were wrong because they weren’t benchmarked against anything that mattered.

This is where the principles in this guide meet a decision most teams make in about ten minutes, a few weeks before launch.

Where most campaign benchmarks come from

Most campaign targets start with a search. Ask Google or ChatGPT for an average engagement rate and you get a figure from a vendor blog, averaged across brands and goals you’ll never see. It says nothing about how you should do against the four or five brands you’re actually fighting for the same buyer.

Last year’s number feels more rigorous because it’s your own. But if your category grew 15% and you grew 10%, you’d hit target and still lose share.

The other usual sources have the same gap. The platform’s own report is marking its own homework. Nobody can check the agency’s “our other clients usually see…”. A scroll through a rival’s ad library shows what they’re running, not whether it works. And a reach target back-solved from finance’s revenue number just multiplies borrowed guesses together.

None of them tells you where you stand against your competitors, or what any of it means for revenue.

Three brands that hit their KPIs and still lost

Getting the benchmark wrong isn’t a rounding error. In three of the best-known cases, the KPI said success while the business went the other way.

• Adidas measured its marketing largely on last-click attribution, so around 77% of the budget went to performance and 23% to brand. When it finally measured properly, brand activity turned out to be driving 65% of sales across wholesale, retail and ecommerce. Simon Peel, its global media director, told the IPA in 2019 that it had over-invested in digital as a result.
• eBay’s paid search reports said ads on the word “eBay” were bringing in sales. Its economists switched them off and watched the same shoppers click the free organic link instead. The KPI had been counting customers who were already on their way.
• Pepsi’s Refresh Project drew more than 80 million votes in 2010. Judged on engagement, it was a triumph. Over the same period Pepsi lost share and slipped to third behind Coke and Diet Coke.

In each case the benchmark came from inside the thing being measured, and nobody was looking at the market or the till.

Better tools, same blind spot

Brand trackers at least compare you with a category norm, but a quarterly survey reports a dip weeks after the campaign that caused it. Marketing mix modeling ties spend to sales, months after the decision. Holdout tests read a single campaign well, but slowly, and they say nothing about competitors.

The newest arrivals, neural networks and double machine learning, try to pull cause and effect out of ordinary campaign data with no control group. When Brett Gordon and Florian Zettelmeyer at Kellogg tested that idea with Meta’s researchers on 663 Facebook campaigns where the true answer was known, the models couldn’t reliably find it. For purchases, the real median lift was 6% and the model put it at 24%. Prediction and causation are different jobs.

Short KPIs miss most of the effect

The 2026 CMO Survey found 70.6% of marketing leaders moving toward short-term impact, pushed there by CEOs, boards and CFOs. The same survey found marketing’s effect on customers now lasts up to six months. So teams get judged over a few weeks on work that pays back over half a year. A KPI borrowed from last year and checked when the flight ends will miss most of it (see Chapter 3).

What a good campaign KPI looks like

Every principle in this guide applies to a single campaign target.

• Benchmarked against the competitive set. Same category, same shoppers, same weeks. Competitors won’t fill in your brand survey, but they turn up in the same search, scanner and ad spend data you do.
• Built on what people do. Searches, visits, views, mentions, what goes through the till. Behavior can be checked against sales.
• Wider than the channel the money went into, AI included. A TikTok push turns up in branded search a week later, or in which brand ChatGPT recommends for a vitamin C serum. If AI assistants are shortlisting brands in your category, share of AI recommendations belongs on the campaign scorecard next to share of search (see Chapter 7).
• Connected to revenue while there’s still time to act. Agreed before launch, read while the campaign is still running, tied to sales. Three months later, you’re reading a post-mortem.

Setting the target in the first place

Once you can see where you stand against your competitors before anything goes live, the target gets much easier to set. The goal stops being “last year plus 15%” and becomes something like “close the gap to the category average on visibility this quarter,” with an estimate of what that’s worth in sales next to it.

The team can steer by that while the campaign runs. And when it goes to the boardroom, the CFO can see the money attached.

What to do this quarter

1. Write down where each current KPI came from. Industry average, last year, the platform, the agency or back-solved. If none of them involves a competitor, that’s the gap.
2. Re-base one live campaign against the category. Compare your growth with the category’s over the same weeks. Share of search is a free place to start.
3. Add AI visibility to the campaign scorecard. Ask the main assistants your buyer’s questions before launch and again after, and note how often you’re recommended against your rivals.
4. Agree the targets and the read-out date with finance before launch. Read it mid-flight, not three months later.

If you’re a CMO: before the next launch, ask one question of every KPI on the plan: compared with whom?

Hitting your KPIs only counts if the brands next to you on the shelf didn’t hit theirs harder.

Chapter 9: Workbook — Your 30-day measurement audit

This is where the principles turn into a plan. Run the audit over four weeks, score yourself honestly, and you’ll know exactly where your measurement is strong and where it’s leaving you blind.

Week 1: Audit the wall

List every measurement tool, report and study you pay for. For each, note the question it answers and how late the answer arrives.

Then count the hours. Estimate how much of your team’s month goes on reconciling sources rather than making decisions.

Week 2: Define your competitive set

List the five to eight brands your buyer actually compares you with: on the shelf, in the same search results, in the same AI answers. Then pull share of search for all of them over the last 24 months. It’s free, and it’s often the most revealing chart a team has seen in years.

Week 3: Talk to finance

Sit down with your CFO or finance lead. Ask two questions: what evidence would make brand spend defensible, and what number do you currently distrust most? Write the answers down. They’re your brief.

Week 4: Score yourself

Score each statement from 1 (not at all) to 5 (fully true today).

What your score means

The four questions to keep asking

Whatever your score, these are the questions good measurement should answer every month:

• Are we gaining or losing momentum against the brands we compete with?
• What’s driving it?
• What happens to sales if we hold spend, or cut it?
• What should we do next, while there’s still time to act?

Where Maitrics fits

We founded Maitrics because we kept seeing the same thing at challenger brands: good marketers making big decisions on instinct, because their measurement couldn’t give them a clear answer in time.

Maitrics is always-on behavioral brand intelligence, built to fill the empty corner from Chapter 3. Our Brand Momentum Index (patent pending) measures all marketing activity across Paid, Owned, Earned and Search, including AI assistant visibility, against your direct competitors. It’s built on third-party market data, so there’s nothing your team needs to feed it. And Eva, your always-on analyst, answers questions in plain language, in the same shape every time: what’s happening, what it means, what to do next.

It isn’t a replacement for the tools you already trust. Keep your MMM for verification and your tracker for the why. Maitrics is the monitoring layer that connects them, benchmarked against your competitors and tied to revenue.

In our own back-testing, we hid the sales data, retrained the model blind and asked it to forecast. It was roughly 93% accurate week by week for a leading condiments brand, and around 91% month by month for a prestige fragrance brand, against 70–77% for naive baselines. That’s out-of-sample prediction, not proof of cause. We think that honesty matters.

If your audit score surprised you, we’d be glad to show you where your brand stands against your competitors. Let’s chat.

Your brand. Quantified.