Brands that keep the same creative team make better work, and we have an obvious interest in telling you that.

The line on our Strategy page reads: "A deck won't grow your brand. Continuity will." We wrote that from experience and have been repeating it for a while without pointing at anything. There is a large, careful study that says the same thing. We found it this week, which is late, and the honest way to use it is to put our own stake in it on the table first. A production company arguing that you should stay with your production company is not a neutral party. Read the method, not us.

The study

Compound Creativity, from System1 with data from the IPA's Effectiveness Databank. It looked at more than 4,000 ads from 56 brands across 44 categories, over five years, representing £3.3 billion in television advertising spend, with responses from 600,000 people. Emotional response came from System1's Test Your Ad platform, brand effects from the IPA and YouGov, and business outcomes from the IPA Effectiveness Databank, which holds more than 1,500 award entries going back forty years.

It scored each brand on thirteen consistency features and built those into a Creative Consistency Score. The features group into three: the foundations (brand positioning, creative idea, how long the brand has been with its creative agency), the culture around the work (letting ads wear in, staying consistent across channels, reusing assets, committing to entertaining people), and the execution itself (brand assets, tone of voice, slogan, characters, soundtrack).

Then it compared the most consistent brands against the least. The most consistent averaged a 3.3 Star Rating, which System1 reads as good long-term brand-building potential. The least consistent averaged 2.6. Over the five years, the most consistent gained 0.2 Stars a year. The IPA's original release puts the least consistent at no growth over the same period. System1's own later write-up of the same study puts them at a 0.3 Star decline instead. We are using the IPA's figure here, the earlier and more conservative of the two, but the gap between the sources is real and worth knowing before you go looking at the study yourself. By the end the study expects the consistent brands to grow market share more than twice as effectively on the same media spend.

The business end is where it gets serious. The most consistent brands generated 27 per cent more Very Large Brand Effects and 28 per cent more Very Large Business Effects, which in the IPA's terms covers sales value, profit, and market share. The study puts the cost of inconsistency to the industry at £3.47 billion over five years.

And the finding this piece is really about: brands that kept the same creative agency across a five-year period produced higher creative quality and grew more distinctive. Letting ads run long enough to wear in mattered too, the ones replaced early did worse than the ones left alone.

The date

Compound Creativity was published on 9 October 2024. System1 wrote it up again in June 2025.

It is currently circulating inside 2026 commentary about a creativity crisis, usually with no date attached, which turns a two-year-old validation study into this year's news. Two years on, nobody has produced anything that contradicts it, which counts for more than a fresh headline would. But if you see the four-times-profit figure or the consistency numbers quoted at you in a deck this quarter, ask when the study ran.

What we think this means

Three things, and the second two are the ones that cost us something to say.

The mechanism is wear-in, and almost nobody gives work time to work. The instinct inside a business is that a campaign is tired long before the audience thinks so. System1's founder put it as "familiarity breeds contentment", which is a deliberate inversion of the thing everyone assumes. The brands in the study that reused assets, kept the same characters and let ads run were the ones whose creative quality climbed year on year. The ones that refreshed constantly stayed flat. If you have ever killed something that was still working because the team was bored of it, that is the finding, and it has a number on it now.

The evidence is about television, and our clients are not buying television. £3.3 billion in TV spend is not the world a 4WD accessories brand or a golf club lives in. The media weight is not transferable and we are not going to pretend it is. What does transfer is narrower and we think it holds: the same face, the same voice, the same visual language and the same idea, carried across a run of work rather than restarted each time, compounds. That is a claim about how recognition is built, not about how much you spend building it. The study is the best available evidence for it. It is not proof for our clients' circumstances, and anyone telling you a UK television dataset proves anything about your YouTube channel is selling.

We are not an agency of record, and the study's agency-tenure finding is not automatically ours to claim. It measures brands staying with the agency that makes their advertising. We make films and we run channels, which is a different seat. The part we do claim is the part we can point at: work made by people who already know the brand starts further along, and the second year of a relationship produces sharper work than the first, because nobody is spending the first month relearning what the business does. That is our experience and the study is consistent with it. Those are two different levels of evidence and we would rather say which is which.

So the line stays on our Strategy page. It now has a source under it, a date on the source, and a note about where the evidence stops.