Founders come with loud voices. Through the process of refining their elevator pitch until it is as a diamond, they develop a battle-tested way of discussing their companies that no one else is going to be able to replicate.
As companies grow, founders themselves and the people around them begin to wonder exactly how to deploy this.
I see a temptation to run toward one of two poles.
One: the founder, despite still very much actively running the company, retreats into something like an emeritus mode — desperate to show they’ve shaken off the garage, where a handful of people did several roles each. Institutionalization leads to a plurality of voices. Founders emerge reluctantly, for perhaps only the largest showpieces and most prestigious platforms.
Or: they hang on, trying to do everything, even as the number of communications needs starts to multiply. They’ve been used to handing out their number to every interested reporter or investor. And well — that’s admirable, but things approach a limit. First of all and expanding team means increasing numbers of domains that exceed even the knowledge base of the smartest CEO. That is of course why founders stock their business with the smartest names in sales, operations, client relations and product.
The answer (somewhat pat, but we need to say it to get to the interesting stuff) is to find a way to more deeply understand what is special about your founder, so it can be sprinkled into a deeper communications programme.
One way, and this has been the traditional way, is a process I call marinating. Whether it’s a chief of staff or, even better, a chief communications officer, sits in and really spends the time to learn the nuance of the person. Or even better, do what my first boss Tom Walek did: bring back document after document for detailed drafts and corrections, over and over and over and over. I learned what he liked and what he didn’t. An internal sensor.
Now there’s another way.
It’s an idea that should be familiar to many people in fintech: take what was previously unstructured data and turn it into something that can be analysed.
Founders create a corpus. Speeches, documents, decks, emails. Frankly, I think things that are contemporaneous and somewhat spontaneous are better — though much of what’s done in the early years still has that crackle, that verve.
Find this and catalogue it. The right model can certainly help, but things are hiding in the recesses of the internet. YouTube videos – panels, Q&As, conference footage –are gold mines. Generating transcripts is now relatively simple: with a couple of free tools you can start to mine material that’s been sitting idle on the internet for years.
Now you have the raw material, and it’s time to analyse it. Three main axes:
- Emotion and feel — Mark where the temperature rises: the topic that makes them interrupt the interviewer, the anecdote they’ve told four times, the question that visibly bores them.
- Sentence structure and word choice — Count the clauses: do they build long subordinated arguments or fire short declaratives, do they reach for the Anglo-Saxon word or the Latinate one, and which two or three verbs can they not stop using.
- Selective use of data — Note which numbers they cite from memory and which kinds of proof they never touch; a founder who quotes retention figures and ignores headcount is telling you what they think the company is.
Let me be clear: I’ve built what I think is the definitive version of this over the past year, and it’s something available to select Cognito clients. That said, there’s no amazing moat here, and frankly I’m curious what other people are doing with this.
Now, the biggest warning: please, please, please do not create a “CEO bot” and have it generate material for the company. That is very dumb and will not work.
Instead, think of it more like spell check. How is this on brand for us? And, for each individual piece of writing, how much of the founder do we want to come through? This can be a dial, turned up and down as needed. An S-1 may need to start with a deep, passionate letter about the journey so far. What comes later can benefit from a more corporate voice.
And of course, great minds aren’t static objects. Companies are on a journey, and you don’t want everything to sound like it was created by a forked version of the founder from October 2024. Be ready to adjust.
Figuring out exactly what makes a founder tick in prose enables amplification on an order of magnitude far greater than any loudspeaker or bullhorn.
