The book and the show are free, and they stay free. This is the other thing, said plainly: the two of us take on a small number of clients directly. Not a product, not a platform — us, in the room. It is paid, it is hands-on, and it is not for everyone.
AI business transformationA fractional chief AI officer or operating partner, accountable alongside you for whether anything actually changes — not a pilot, not a workshop that ends on Friday.
Most AI programmes buy seats, run training and measure adoption, then wonder why usage went up and nothing else did. The gap is never a production deficit. It is the seams: work that answers a question nobody is asking anymore, ceremonies built to compensate for problems the machines already solved, and a middle layer still routing status that now routes itself.
That is an operating problem, and it is fixed the way operating problems have always been fixed — by someone senior enough to decide, close enough to the work to see it, and around long enough to be wrong in front of you.
Two futures are being built with this technology. They do not differ in their models. They differ in their verb.
Replace
Takes the producing, then asks what the people were for. One company deleted seven hundred agents’ worth of work — and spent the next year hiring humans back.
Empower
Takes the producing and hands the people the four questions. Same technology, same job — one company promoted 8,500 people into a billion-euro channel.
Only one of the two was rehiring.
Fractional chief AI officer and operating partner: in the seat with you for whether anything actually changes. The diagnostic, the seams, the briefing practice, and the middle layer that now routes itself.
Fractional chief AI officer · Operating partner
Fractional CMO and go-to-market advisor, and an executive coach. Positioning, narrative, and the gap between what a company says and what it actually sells — plus the leaders carrying it.
Fractional CMO · Go-to-market advisory · Executive coaching
The engagements below are the AI transformation work, which the two of us run together. Jake’s go-to-market and coaching work is hired on its own terms — ask and we will point you at the right one.
None of this began as a service. Each one started as something we got wrong at our own company and had to fix; the book is the write-up. The chapter references are where each is described in full — free, and without us.
Seven domains, twenty-one points, one honest hour — clarity, ownership, cadence, velocity, load, feedback, growth.
What you get A scored baseline, and the weakest domain named. It tends to surprise the leaders who were most confident.
Written up in chs. 7, 14, 16
Your top project’s leads in one room, each answering the same four questions about the same project. Then we read the differences aloud.
What you get The misalignment found in week two instead of at the QBR, while straightening it still costs one conversation.
Written up in chs. 4, 7
Every recurring meeting on the team calendar, tested with one question: which part of this is memory, and which part is steering?
What you get Automate the memory, keep the steering. Most teams get back about a day a week and miss none of it.
Written up in ch. 12
Train the org to brief machines the way it would brief a new hire: intent, context, constraints, definition of done, staffing and budget by stakes.
What you get Fewer redos, a running cost you can actually see, and the five briefing sins out of the team’s habits.
Written up in chs. 10–11
The relay half of the manager’s job is being automated on schedule. We retrain the middle for the half that isn’t.
What you get Managers who coach instead of route, and the time protected for them to do it.
Written up in ch. 17
Your work already leaves a continuous record. We wire it into a weekly read of whether daily decisions still touch the strategy.
What you get Drift read as a rate, week to week, instead of discovered as an event at the QBR.
Written up in chs. 15, 16
Delivered by Erkang Zheng and Jake Sorofman, directly. Erkang Zheng also runs Ariso, which is a separate thing and not what you are buying here. We take a small number of engagements at a time; the diagnostic is the way in.
Erkang Zheng founded JupiterOne and took it to a billion-dollar valuation without shaking an investor’s hand — and had the worst quarter of his career while every team hit every milestone, because three of them were rowing in three directions.
That quarter is chapter seven. The fix is chapter fifteen. The engagements on this page are what he wishes someone had run on his own company in February instead of April.
Read the book first →Everyone did their job. The jobs no longer added up. The boat spun in a slow, expensive circle.
Before anything is scoped or invoiced, we spend 45 minutes on your situation. You talk; we ask the questions from the diagnostic. It costs nothing and there is no deck.
Come with three things: what you have rolled out, what it was supposed to change, and what actually changed. That is enough for us to tell you where the gap is — and, often enough, that you do not need us. We say so when it is true. The book’s tools are free and most teams can run them without help.
Scheduling link coming shortly — email reaches us in the meantime.