Most founders believe their managers run good one-on-ones. Ask to see last month's notes, the commitments made in them, and what happened to those commitments. The belief usually goes quiet.
Before Edge8 and the AI Officer Institute, I spent years at TINYpulse working with more than 1,000 businesses. Fortune 100 companies like Fidelity, Alaska Airlines and the Boston Red Sox, and hundreds you have never heard of. Their one-on-one coaching programs failed the same way almost everywhere.
Three Reasons, Every Time
- No prep. Preparation takes time, so it does not happen. The manager walks in with whatever is in their head from the last two hours, not the last two months.
- No accountability. Nobody writes down what was promised, with an owner and a date. So the next meeting cannot open with “you said you would.” It opens with “how are things going,” and the hour becomes a catch-up.
- No usable record. A few bullets in a notebook will never show you a trend across six meetings. And you cannot coach a manager on how they coach when there is no record of how they coached.
To be clear, those are my observations from the table, not research findings. But I saw the pattern often enough that I stopped treating it as three problems. It is one problem with three symptoms: nobody owns the record.
Why a Founder Should Care
Gallup found that managers account for at least 70% of the variance in employee engagement scores across business units. Not pay, not perks. The manager.
Gallup's later work on feedback puts a clock on it. Among employees who received meaningful feedback in the past week, 80% are fully engaged. People who get daily feedback from their manager are 3.6 times as likely to feel motivated to do outstanding work as people who get it once a year. Feedback is a weekly discipline, or it barely works.
Now the part most managers never hear. Kluger and DeNisi reviewed 607 effect sizes from feedback studies. On average feedback helps, with an effect size of 0.41. But in more than a third of cases, it made performance worse. Generic, late, or aimed at the person instead of the work, and it does damage.
The job is not more feedback. It is feedback that is specific to the person, tied to what they committed to, and reviewed by a human before it lands.
Those three requirements map one to one onto the three failures. No prep makes feedback generic. No accountability makes it untethered. No record means nobody can tell whether it helped.
Why AI Changes the Math
Every fix I saw at TINYpulse, a prep template, a shared doc, a quarterly review, attacked one failure and added work for the manager. The added work is why none of them stuck.
An AI agent is the first tool I have seen that removes all three without adding work. It does the three things a manager will not do at 7:45 on a Tuesday: read the whole record, remember every commitment, and write documentation clear enough to show a trend.
If you want the build itself, I covered it in how to assemble your own AI leadership coach. This post is the rollout: what to do, in what order, and what it honestly looked like when we did it at Edge8, my company.
The Eight Weeks, One Move Each
Do them in order. Each week's output is the next week's input. The ladder alternates on purpose: weeks one, five and eight fix the record, weeks two, four and seven fix accountability, and weeks three and six fix prep.
Week 1: One Row per Person
Fixes the record. Put each person's goals, priorities and last three meetings on a single record. One row you can point an agent at. Our mistake: we did not get everyone on it on day one, and every person missing in week one became a meeting with no prep in week three.
Week 2: Write Down the Commitments
Fixes accountability. Every promise from every one-on-one gets a line: what, who owns it, by when. Both directions, so the manager's promises go in too. The log will not make people follow through. It makes not following through a visible fact instead of a vague feeling.
Week 3: Let the Agent Write the Prep
Fixes prep. Before each meeting, the agent writes the manager a prep from the record only, nothing invented, and the manager reads it in two minutes. When we had the prep, the first ten minutes stopped being small talk and started on the open commitments and the goal that had slipped.
Week 4: Share the Agenda
Fixes accountability. Both sides add talking points before the meeting. It becomes their meeting, not a status report to a boss. We built the feature and forgot to build the habit.
Week 5: Capture the Transcript, Draft the Recap
Fixes the record. Record the meeting and let the agent draft a two-tier recap: a candid manager-only tier and a shared tier for the person. The manager reviews before anything publishes. Build a manual paste fallback first. Our transcript pipeline broke for a month, and without a fallback that is a month of coaching that never existed on paper.
Week 6: Profile the Person
Fixes prep. Publish an OCEAN profile, a five-factor personality profile, so the prep and the recap can adapt tone. This is the week that most directly protects against the third of feedback that does harm.
Week 7: Check In Mid-Cycle
Fixes accountability. Between meetings, a one-minute prompt: what moved, what is stuck, what do you want to talk about. It feeds the next prep, so the meeting starts where the person actually is, not where they were two weeks ago.
Week 8: Read the Trends and Coach the Coach
Fixes the record. Now the record can teach. A trend report per person, the commitment follow-through rate, and a mode split per meeting: coach (asking), mentor (advising) or direct (telling), against a deliberate 80/15/5 target. That split is the first honest picture most managers ever get of how they coach.
Our Scoreboard, Gaps Included
Here are our numbers after six weeks on the cycle at Edge8, including the ones I would rather not print.
30 one-on-ones held. 10 with an AI-written prep. 12 with a transcript captured. 14 with a published recap. 57 commitments logged: 19 completed, 4 on track, 29 open, 2 blocked, 2 dropped, 1 needing attention. Behind those: 14 people coached, 4 published OCEAN profiles, 5 trend reports, 10 check-ins, 4 shared talking points, and nine surveys with 332 responses feeding the same records.
What worked: when the prep existed, the meeting started on substance, and the commitments log turned follow-up from a memory into a fact.
What did not: two thirds of meetings ran without a prep. Fewer than a third of the team has an OCEAN profile. Check-ins are rare. And the model's first recaps praised more than they reported.
I am not claiming engagement or performance improved. We have not measured that yet. I am claiming the three failures are now on a screen with a number next to each, and a thing you can see is a thing you can fix.
The Review Gate Is Where the Leadership Lives
Can the recap go straight to the employee? No. Our model's first recaps were warm, generous and light on what actually happened. Exactly the kind of feedback that lands in Kluger and DeNisi's harmful third.
The review takes the manager a few minutes. It is where they decide what the person needs to hear, in what words, and what stays in the manager-only tier. That is the coaching. The agent removes the preparation, the remembering and the writing. It does not remove the judgment. Take out the review gate and you have automated the feedback that hurts.
What to Expect From a Certified Leader
Every rung on the ladder assumes someone owns the record and the cycle: the schema, the roll-forward, the pipeline, the fallback when it breaks, the review gate, and the awkward question of which manager is telling when the target says they should be asking.
At TINYpulse, that person did not exist. Managers do not own records. HR owns forms. IT owns tools. Nobody owns the thing in the middle, and that is why the three failures kept coming back.
So as a founder, I treat coaching with AI as a leadership skill, not an HR program. A leader I would trust with a team should be able to stand up the record, run the cycle without skipping the review gate, and read a scoreboard like ours without flinching or spinning it. It is the same judgment I described in the three skills it takes to lead AI: knowing what good looks like, and owning the system that produces it.
Start This Week
Do week one in the next five working days. One row per person, with their goals, their priorities and the last three meetings. No agent required yet. Just the record. Then count how many of your people have that row today. If the answer is none, the fix starts with deciding who owns it.
That decision is the hard part, and it is the one AI Officer certification is built to inform. Your leaders work through core challenges on your own data and workflows, attend four live coaching sessions, and within 90 days you know who is best equipped to lead. Find out which of your leaders can do this.