Our team recently took an hour to discuss why we are integrating AI into Honeycomb, and get the feelings around that onto the table before we dive too far in. I have sat through enough transformation programmes that lived and died in PowerPoint to believe in the value of this sort of conversation.
I know a lot of firm leaders are in the same place we are - convinced AI matters and unsure how to bring a team with them without either forcing it or letting it sprawl. So here is the conversation we had, in case it is useful before you have yours.
Start with the why
We started with the ‘why’. For me: we have grown quickly over the last few years and we want to keep growing, but in a way that stays lean and keeps the close-knit team we have, rather than adding heads to match every rise in demand. Done well, AI is what makes that possible. It also keeps us a credible step ahead of clients, who increasingly ask how we are using it. A training firm that is a laggard on AI is a hard sell, and some of the conversations I most look forward to now are the ones where we sit with a client and work out what this technology can really do for a firm like theirs.
The team added other reasons. One was consistency. Good AI tooling forces you to decide exactly how you do things and what “good” actually looks like, and in a premium business where quality is critical, that discipline is a big prize. Another was the founder bottleneck. If the team can produce more complete, more considered first drafts, senior review becomes lighter and the work stops queuing behind two or three people. A third was innovation: the chance to design training that does not feel stale, and to stand out by doing things most of our market has not thought of yet.
The harder question
Then I asked the harder question. Not “is this a good idea?”, but “how do you actually feel about it?”.
The primary worry that came up was that speed trains you to value getting something done over thinking it through. Work looks polished even if it hasn’t been thought through, and it becomes harder to say where “good enough” sits.
Several people described a particular kind of mental fatigue: working at the machine’s pace while your brain still runs at human pace. The ideas that used to arrive on a dog walk, in the shower, or in the car now get squeezed out by the rush to the next task. One person said they finish some days feeling tumbled about, three tabs open, moving fast but not processing. Another said the tasks never seem to finish, because the moment the thing is done you roll straight into improving the process that made it, and fifteen minutes easily becomes an hour.
There was also a sense of loss, which I recognised myself. AI can now do things I used to be proud of being able to do, and do them quickly. For a while that frustrated me. Somewhere along the way it turned into something closer to feeling empowered: proud of how much I can do well, and how quickly I can respond to clients. We all seemed to share a love-hate relationship with Claude, excited by the possibilities while watching it take over parts of the job we enjoyed.
I think that ambivalence is the right place to be. The people who are purely evangelical worry me as much as the people who are purely resistant.
The fears help us understand how to do it well
The fears are not arguments against doing this, but perhaps a specification for how to do it well. The second half of our conversation was about that.
The first decision was to treat this as a proper programme with its own structure and owners, rather than something squeezed in around delivery. We had let things run a little wild, and while I would rather that than paralysis, it was not sustainable. So, we’ve now mapped our whole business as a set of connected processes, from a client first becoming aware of us through to delivery and beyond, and we are working through each one carefully, deciding what the tool should do, what it should not, and where a human judgement has to sit. Each process gets a clear owner - the person accountable for the outcome, who may or may not be the one doing the hands-on work.
We also broke the work into smaller chunks than our usual quarterly goals, so progress stays visible and nobody disappears into a solo build for three months. And we built in some deliberate brakes against the fears the team had voiced.
Quality matters more than speed, because we are setting the business up for the future, so we are explicitly telling ourselves not to rush it at the expense of doing it well. The work is collaborative by design, so it does not become one person alone at a screen. And my own strong advice to the team was to start with a pen and paper, or at least with your own thinking, before opening the AI at all. The tool does not yet know how we work. If you let it lead, it will hand you back something confident-sounding, but it may be missing some important context and nuance.
My advice? Make room for the conversation
Our direction of travel is better as a result of that hour. It was important for people to say it out loud: they love it and resent it, they are faster and less certain. Left unsaid, none of that goes away, it just shows up in unhelpful ways further down the line.
So the recommendation I would make is book the hour. Give your own answer to the why first, then ask how people actually feel, and take on board what comes back rather than reassuring it away.
I believe the firms that get the most from AI will be the ones most deliberate about protecting judgement while everything else speeds up, and that judgement sits with the people in your team.
How is your team really feeling about AI? That conversation might be the most useful one you have this quarter. If you would like to compare notes, drop me a line.
Thank you for reading The Skilled Consultant. If you haven’t yet subscribed, please do so to receive all our articles direct to your inbox.
There are several other ways you can interact with Honeycomb Consulting Skills Training….
Connect with Deri Hughes (Founder & MD) on LinkedIn
Connect with Colin Mann (MD) on LinkedIn
Book a 30 minute intro call with Deri Hughes
Book a 30 minute intro call with Colin Mann
Stay informed about our free workshops and webinars - follow Honeycomb on LinkedIn or visit our website.
Making room for the conversation around AI is important - but it begs a different question. How much room is there truly? As organizations increasingly incorporate AI capabilities across enterprise, is the question of its implementation inevitable already, or are there truly discussions taking place about where and how it is rationalized, and more importantly, whether there are areas in the business or organization where it is suboptimal to deploy at the current time?