Founder note
What 15 years taught me about creative that sells
The work that sells and the work that wins awards are not enemies. But only one of them pays the bills.
I have been doing this for fifteen years now, across more markets than I expected to ever see, and I have changed my mind about almost everything except one thing. The work that sells and the work that wins awards are not enemies. They are not even rivals. But only one of them pays the bills, and confusing the two is the most expensive mistake I have watched smart people make, including me, more than once.
What follows is not a theory. It is a short list of things I had to learn the slow way.
A few things I had to learn the slow way
The first is that work starts with a real human truth, or it does not start at all. Not a brand truth, not a positioning statement, not a clever line. A real thing a real person feels before they have any reason to care about you. When the work names something true, people lean in. When it does not, no amount of production saves it. Early on I mistook surprise for truth. They are not the same. Surprise wears off in a week. Truth keeps working.
The second is that distinctiveness is an asset you protect, not a style you chase. The brands that grew under our care were the ones that looked and sounded unmistakably like themselves, in the third year as much as the first. The temptation is always to refresh, to follow the new thing, to redo the work because the team is bored. The audience is not bored. They are barely paying attention. Consistency is not the safe option. It is the compounding one.
The third took me longest. You answer for the result, not the deliverable. For years I thought my job ended when the film was beautiful and the client was happy in the room. It does not. The job ends when something moved in the market. A campaign that everyone admired and nothing happened after is a failure with good manners. I would rather make a less perfect thing that shifts a number than a flawless thing that shifts a mood.
A campaign that everyone admired and nothing happened after is a failure with good manners.
The fourth is that scale is a craft, not a compromise. Making one beautiful film is hard. Making the idea hold across forty cuts, ten languages and three markets without going limp is harder, and it is where most good ideas quietly die. We learned to treat the hundredth version with the same care as the first, because the customer who sees the hundredth does not know or care that they got the tired one. To them, it is the brand.
What has not changed in fifteen years
The idea still decides everything. I have sat through every wave of new technology, new format, new platform, new measurement model, and at the end of each one the same thing is true. A sharp idea built on a human truth moves people, and people move numbers. A weak idea with a big budget moves nothing, slowly. The brief, the tool, the media plan, the cast, the edit, all of it exists to protect the idea, not to rescue it. If the idea is wrong, there is nothing to protect, and everyone downstream is just being expensive together.
This is the part that stays steady while everything else shifts under it. It is the reason I trust craft over fashion, and the reason I have never believed that the newest tool is the same as the best answer. The tools change the making. They do not change the deciding.
What AI did change
AI changed the cost and the speed of making, and almost nothing about the standard. That distinction matters more than the headlines suggest. We can now put a small team at the output of a large one. We can try ten versions of an idea in the time it used to take to argue about one. We can make the idea travel into more languages and markets without the budget that used to gate it. All of that is real, and all of it speeds the making, not the thinking.
What AI cannot do is decide what is worth saying. It does not know which truth will land, which line is brave enough, which cut to kill. That judgement is still human, still slow in the right way, still the place the work is won or lost. I have written before that AI is a tool, not the idea, and fifteen years tells me that is not a slogan. It is just how the work behaves. The tool got faster. The bar did not drop.
The throughline
If there is a single thread running through all of it, it is this. Creative work is not in the business of being liked. It is in the business of moving a real person to do a real thing, and being honest enough to measure whether it did. The truth feeds the idea. The idea earns distinctiveness. Distinctiveness compounds. And the only fair test, in any year, with any tool, is what changed in the market afterwards. That is the whole of it, and it is also the part I wrote about more fully in creativity that grows the business.
Fifteen years in, I am more certain of less, and the less I am certain of is sturdier than anything I believed at the start. Start with a truth. Protect the idea. Answer for the result. Everything else is craft and timing and luck. Those three are the work.
Questions, answered
What makes creative work actually sell?
It sells when it is built on a real human truth, makes the brand instantly recognisable, and is judged by the business result rather than the polish of the deliverable. The idea has to be distinctive and useful to a real person making a real decision. Charm without a job to do is decoration.
What is the biggest lesson from 15 years in advertising?
The idea decides everything. Budgets, tools and timelines change, but a sharp idea built on a human truth is what moves people and moves numbers. Everything downstream, including production and media, exists to protect that idea, not to rescue it.
Has AI changed what makes creative effective?
AI changed the cost and speed of making work, not the standard it is held to. It lets a small team produce at the scale of a large one and test more ideas faster. But it cannot decide what is worth saying. The judgement is still human, and that is still where the work is won or lost.
