How AI agents cut discovery from months to days at Aidols Group
Sepehr Sisakht is president of Aidols Group, a consulting firm that builds AI agents to interview every team, map how work actually happens, and deliver production-ready systems rather than strategy decks. The company guarantees measurable ROI with fixed fees, and its use of AI to run discovery interviews has shortened the discovery phase from months to days while producing more candid data.
Hand discovery interviews over to AI agents: they can gather more honest, structured operational data far faster than traditional consulting, which lets teams ship targeted, measurable AI systems quickly – while keeping humans responsible for the final, judgment-based recommendations.
What’s the quick origin story of your business, and what makes what you do genuinely different from the other options your customers are looking at?
AIDOLS started in 2024 out of a frustration we kept seeing on the client side: companies were making big AI bets based on a blurry picture of their own operations. Traditional consultants would run 20 interviews, filter everything through hierarchy and politics, and hand over a strategy deck six months and seven figures later. The deck sat on a shelf. So we built the thing we wished existed, AI agents that interview every team in an organization, map how work actually happens (not how the org chart says it happens), and surface the two or three places where AI creates real leverage. Then we build and ship those systems in weeks, not quarters. The structural difference is that we’re engineers, not analysts. Our deliverable is production code and an operational map, not slides. And we put our money where our mouth is: fixed fees, and if the AI doesn’t deliver measurable ROI, you don’t pay. No one else in the consulting space is willing to sign up for that. For reference, our wesite is aidolsgroup.com
Where does most of your new business come from today, and where do you wish it came from?
Most of it today comes from referrals and inbound through our content, research reports, case studies, and honestly, ranking for the questions operators are already Googling. The ROI guarantee does a lot of heavy lifting too; it gets forwarded around leadership teams. Where I wish more came from: mid-market operators who’ve already been burned by a failed AI pilot. Those are our best clients because they’ve learned the expensive lesson that the model isn’t the problem, the missing operational map is. Right now they tend to find us second, after a bad experience, rather than first.
What’s one thing you’ve actually handed over to AI in the last year, and one thing you tried it on and went back to doing the human way?
Handed over completely: the discovery interviews. Our AI agents now conduct the operational interviews across 100% of a client’s teams. Counterintuitively, people are more candid with an AI than with a consultant taking notes, no career risk in telling the truth to a machine. That single change took our discovery phase from months to days and made the data dramatically better. What we tried and took back: writing the final executive recommendations. We let AI draft them for a stretch, and the output was competent but hedged, it read like every option was equally valid. Clients pay us for judgment: "do these three things, in this order, skip everything else." That call requires a human who’s willing to be wrong. AI now assembles the evidence; a human makes the recommendation.
Have you noticed a change in how customers find you or what they already know before they reach out? Is anyone arriving through AI assistants like ChatGPT yet?
Yes, and it’s the biggest shift we’ve seen in 12 months. Prospects now arrive pre-educated, they’ve already asked ChatGPT or Claude to compare AI consulting approaches before they ever hit our site, so first calls start three steps deeper than they used to. We’re seeing measurable referral traffic from AI assistants, and we know we’re being cited in AI-generated answers because prospects quote our own framing back to us – a compass, not a contractor – before we’ve said it. It’s changed how we write: we now publish content structured to be a good, citable answer to a real question, not just to rank on Google. AI assistants are becoming the new first impression, and most businesses haven’t noticed yet.
If you were advising someone in your industry on all of this for the next 12 months, what would you tell them to actually do, and what would you tell them to ignore?
Do: map your operations before you buy anything. Pick one workflow with a measurable dollar outcome, ship something to production in 90 days, and instrument it so you know whether it actually worked. And demand ROI accountability from anyone you hire, if a vendor won’t tie their fee to a measurable result, that tells you what they believe about their own work. Also, start paying attention to how you show up in AI assistant answers, because that’s where your next customers are forming opinions about you. Ignore: multi-year "AI transformation roadmaps," any engagement whose final deliverable is a PowerPoint, and the model-of-the-week discourse. Which frontier model you use matters far less than whether you understand your own workflows well enough to know where to point it. The companies winning right now aren’t the ones with the best AI, they’re the ones with the clearest map.
Thank you to Sepehr Sisakht and the team at Aidols Group for sharing what actually worked with Leaders Perception readers.
Want to share your playbook?
Leaders Perception publishes founder and operator interviews every week. A few questions, about five minutes, no calls. It is free and everyone we accept gets published.