‘Relentless pursuit of the truth’: Why are marketers dissatisfied with metrics?

You could be excused for thinking something must be rotten in the world of marketing effectiveness measurement.

Data from Marketing Week’s 2025 Language of Effectiveness survey, in partnership with Kantar and Google, shows nearly half of marketers are unhappy with the analytics available to them, a figure that rises to more than half of CMOs.

Perhaps just as worrying is that more than a third of respondents admit that they don’t always understand the connection between the analytics they receive and future marketing decisions taken by their brand.

Is half of all of the time and effort expended on measuring effectiveness really wasted?

ICE Travel Group CMO Steve Seddon is responsible for brands such as Icelolly.com and Travelsupermarket. He says there is clearly work to do if marketers are to fully trust the effectiveness figures available to them. The sheer volume of data is one issue to contend with.

“For short-term measurements, from a last-click perspective, it’s good. But I’ve been in this job for two and a half years and it probably took a year and a half for me to feel comfortable with the analytics and the measurement that we’ve got,” he says.

“The more touchpoints you add, the more that you mix online and offline, the more that you’re looking at longer-term activity, the more complicated it gets.”

Marketers probably feel like they’ve got data coming out of their eyeballs, because it’s everywhere, but they struggle with the certainty that they want.

Ross Farquhar, Little Moons 

ICE has worked with econometrics specialist Magic Numbers on marketing mix modelling, which has helped with longer-term metrics, says Seddon. But not all brands have the budget for that kind of service, he concedes.

“We’ve got strong tools at our disposal now as marketers, like better pre-testing than we used to have, with things like System1 and its equivalents, and better AB testing tools. Marketing mix modeling is getting more accessible. But yes, they’re all expensive,” says Seddon. 

In addition to that cost, greater knowledge can also lead to greater scrutiny from outside the marketing team, and take time away from day-to-day marketing tasks, especially at smaller companies. For marketers with limited time and budgets that might beg the question of whether they want to spend limited funds on getting data that could make their jobs more complicated, adds Seddon.

Even larger groups can find dealing with data and analytics is taking up an increasing proportion of their resources.

“You’ve got all these different systems and tech stacks, and they’re growing in strength. They’re going to grow in importance. So you’ve got the CRM team and the tech that they use, and that’s caught within a silo. Then you have the same for PPC or paid social and then the advertising activity we do. Trying to stitch them together has been complicated,” says Seddon.

Another key issue that makes it hard to grasp the nettle of effectiveness is the pace of change in an industry caught in the middle of a drawn out digital revolution, suggests Little Moons marketing, sustainability and innovation director Ross Farquhar. That pace of change shows no sign of slowing down.

“If I want to prove the case for a big national TV campaign, I’ve got reams of data that would do that and I also have the means of testing that through econometrics and other forms of modelling. But increasingly, proportions of budgets aren’t going on big national TV campaigns,” he says.

“They’re going onto platforms like Instagram and TikTok, and a lot of the measures associated with those platforms can be quite short term. They just haven’t built up the data bank of effectiveness cases.” 

Half of CMOs ‘unhappy’ with measurement of creative effectiveness

In addition, the pace of change has upended traditional practices. For example, when he joined Little Moons five years ago, Farquhar was able to make the point the mochi ice cream brand should advertise on TV based on such figures.

“In the last four years, the world has changed so fast that if I stood up with that same data bank now and talked about that to my board, they would just not believe me,” he says.

“All of your stakeholders are people who’ve got kids at home, who they watch infinitely scrolling every single night. They don’t sit at home together watching Coronation Street at 7.30pm on a Wednesday.”

Even an equivalent data bank of the past performance of ads on social platforms would not necessarily predict how content might perform now, he adds.

“The algorithms are constantly changing, so where the data bank is brilliant at telling you best practices and what results have been in the past, that may not bear any resemblance to the moment in which it’s being read,” Farquhar explains. 

“So I can imagine marketers probably feel like they’ve got data coming out of their eyeballs, because it’s everywhere, but they struggle with the certainty that they want or that their stakeholders are demanding. That creates frustration.”

Reaching 80% accuracy

Digital media platforms have a wealth of data, but the sheer volume can heighten the challenge for marketers.

“It might not be the right data to actually understand effectiveness and it can lead you astray,” says Farquhar.

“I’ve had lots of conversations about the efficacy of the engagement rate on TikTok, versus somehow trying to calculate share of voice brand lift studies, to try and get to some kind of sense of the medium-term effects of this sort of stuff. It’s just endless amounts of data.”

Smaller brands will find it hard to match the spend of larger rivals when it comes to investing in bespoke quantitative studies, which cost a similar amount regardless of the size of the client. However, that is not the end of the unfairness in a sector where might is often perceived as right.

“A lot of the data and measurement that comes from those platforms is thrown in for free by the platforms when you spend a certain amount, but a lot of SMEs may not be meeting that threshold,” Farquhar adds. 

“They are faced with the choice of, would I rather spend £20,000 on that brand lift study, or would I rather spend it on more media and just take the leap of faith? Most, I suspect, will take the leap of faith, but feel quite disgruntled about it because it means they don’t have the data and the evidence for its effectiveness.”

As CMO of cyber security brand CyberSentriq, Dryden Geary is in a sector fortunate enough to be awash with data. In fact, a B2B SaaS business, the real challenge can be too much data, he says.

The more touchpoints you add, the more that you mix online and offline, the more that you’re looking at longer-term activity, the more complicated it gets.

Steve Seddon, ICE Travel Group

Despite increased resistance to online cookies and the rigours of GDPR, the group still has access to enough data to make consolidation a challenge. However, when preparing for the merger of TitanHQ and Redstor that formed CyberSentriq in June, the marketing team took a deep dive into the effectiveness figures.

“When you’re in front of a group of investors, you’ve got to have this really at the front of your mind, you’ve got to really understand it. There can’t be any gaps. So we have had chance to really get this right and get our views on this set. From a B2B perspective, there’s loads there,” says Geary.

Amid this data consolidation, he has tried to instil a mindset which says the data and the model “don’t have to be perfect”. 

With a complex B2B sales process and long customer relationships, it is important to keep a strong sales pipeline and maintain the correct direction of travel. On that basis, when measuring effectiveness to source sales leads, a percentage of accuracy in the high 80s is deemed perfectly adequate. Striving for 100% accuracy in marketing would just slow things down, Geary argues. 

“We can operate quite easily in the high 80s from an accuracy perspective, because I’m looking at trends,” he says.

However, getting the highly analytical people who are drawn to SaaS careers to accept accuracy below 99.9% can be a challenge in itself.

“We look at things like Google Search Console. There’s so much stuff in there, it’s just so developed, that we then have our own methodology internally of tracking sales and tracking activity online. We bring everything into a [cloud-based] data warehouse in Snowflake, and we query them in [data management system] Tableau, to consolidate. But if you didn’t have that team with the skillset to do that, you would have these three different data sources,” says Geary.

Tale as old as time

Just like budgets for market mix modelling, experienced data scientists are not distributed equally through companies. Geary also makes the point effectiveness data is essentially historic and what CyberSentriq really wants is a prediction of what activity will be effective in the future.

“Really, you’re living in the quarter coming ahead of you,” he states.

For example, the group will be seeking to predict how many potential customers it might meet at a forthcoming trade show and consolidate that with ongoing marketing activity that is happening online – not a straightforward task.

“How can someone grab something like that, which is a real world occurrence and merge it into the digital activity that’s happening, and then answer the key question for me, which is what is the pipeline for the next quarter?” asks Geary.

Under 40% of marketers always measure ROI

Happily, in the B2B space at least, there are effective options to help answer that question. CRM providers, increasingly with the help of AI, are bringing in new functionality that can analyse and consolidate multiple inputs, and provide effective top line figures that are understandable and actionable, he says.

Of course, the digitisation of data and measurement can never remove the human element of marketing entirely. That may leave some elements of the dissatisfaction with effectiveness metrics open to question, says Farquhar.

Brinkmanship may also be playing a part, he suggests. Marketers always want more for less from their analytics, so will always say that they aren’t happy with them. If the conversation with digital platforms is to advance, clients essentially need to remain dissatisfied with the effectiveness measures they are given, so that they improve.

“This is a tale as old as time, but marketers don’t want to say that they’re completely satisfied with effectiveness data,” he adds. 

“If their CEO or CFO, or board members are not satisfied that they fully understand whether their marketing is working, it isn’t a great look for the marketer to be like: ‘Oh, I’m fully satisfied.’ They want to portray themselves as being in relentless pursuit of the truth.”

Marketing Week will continue its reporting from the Language of Effectiveness research in the coming weeks. Click here to download our report on the data.