The CPO Listening Tour Is Broken. Run a Ledger Instead.

Every new CPO gets handed the same advice: spend your first month on a listening tour. Forty coffee chats, open questions, lots of nodding. I have run that tour. It produces a word cloud. At day 30 you have a stack of notes, a list of themes the org already knew about itself, and no way to tell which of them are true. You have spent a month of your evidence budget on impressions.

The listening is not the problem. The format is. Unstructured conversations reward the most articulate person in the building, not the most accurate one. So I stopped running tours and started running a ledger.

The short version

The CPO listening tour, rebuilt as a listening ledger: ask every interviewee the same nine questions, log every answer as a falsifiable claim with a source, a confidence level, and a field for what contradicts it, then mine the contradictions between functions as the real signal. Run a Friday synthesis to update confidence levels, and hold all conclusions for 30 days so your information supply stays uncurated. This is the month-one deep dive of The CPO 30/60/90, and it borrows the falsifiable-priors discipline that Michael Watkins gestures at and most people skip. The ledger template and the interview script are in the CPO First-90 Kit(coming Jun 10).

The nine questions, verbatim

I ask the same nine questions in every interview, in this order, regardless of who is across the table. The repetition is the point. When forty people answer identical questions, divergence becomes measurable. Here they are as I actually say them.

1. "Walk me through the last decision you were part of that you disagreed with. What happened?" Opens with a story, not an opinion. Stories contain checkable facts.
2. "What should I kill?" Four words. The answers cluster fast, and the things three or more people name independently go straight on the kill list draft. People light up at this question. Nobody has asked them in years.
3. "Where does this org lie to itself?" Phrased exactly like that. The flinch tells you as much as the answer. The people who answer instantly have been rehearsing for someone to ask.
4. "What did my predecessor get right that everyone now dismisses?" New leaders inherit a reflex to discount the previous regime. This question is the antidote, and it surfaces the load-bearing decisions you must not accidentally undo.
5. "What is the eval score on our most important AI feature, and which way is it trending?" The maturity X-ray. Most people cannot answer, and the pattern of who can is your real quality org chart. The reasoning behind this question is in The Eval Is the Spec.
6. "If you had my job for one quarter, what is the first thing you would do?" Everyone has a draft answer. The function-by-function clustering tells you where the pressure is.
7. "Who do you go to when you need the truth about how something is going?" Not who they report to. Who they trust. Map the answers and you have the informal information network, which beats the org chart every time.
8. "What commitment has leadership made and quietly dropped?" This one fills your trust-repair backlog. The org remembers every dropped promise. You are inheriting that debt whether you log it or not.
9. "What did I not ask that I should have?" Standard closer, and it earns its slot. Maybe one in five answers matters, but that one is usually a claim nobody volunteers unprompted.

Notice what is missing: "what are our biggest opportunities" and every other question that invites a prepared narrative. You will get the prepared narrative anyway. Do not spend a question on it.

The ledger schema

Every claim from every interview gets a row. Five fields.

The discipline of writing claims as falsifiable sentences does most of the work. "The sales team oversells the AI features" becomes "deals closed in the last two quarters included commitments the product cannot meet, and I can check the last ten contracts." One is a mood. The other is a task.

Contradictions are the product

By interview fifteen, the contradictions field starts to glow. Sales says the big account churned because engineering shipped late. Engineering says the deal was sold on features that never existed. Customer success says the account was never onboarded properly and both other stories are downstream of that. Three functions, one event, three causal stories.

This is not noise to be averaged away. This is the highest-value output of the entire month. The gap between the stories maps the real operating model: where information dies, who controls which narrative, and which function's version of reality the CEO has been buying. Agreement across functions is usually just the official story echoing. Disagreement is information.

I keep a separate contradiction list, sorted by stakes. The top three contradictions become the first things I resolve with actual evidence in month two, and resolving them publicly is worth more credibility than any quick win. The instrumentation that settles them is the subject of CPO days 31 to 60(coming Jun 15).

The Friday synthesis

Every Friday, ninety minutes, calendar-blocked, no exceptions. Four steps.

First, read every new row from the week. Second, update confidence levels, but only where new evidence arrived, never because a claim got repeated. Third, refresh the contradiction list and pick which contradictions are now resolvable. Fourth, reread your day-zero priors file and mark which priors the week's evidence supported, broke, or left untouched.

That fourth step is the one people skip and the one that matters most. Your priors are claims too. They go in the ledger with you as the source, and they get falsified on the same terms as everyone else's. I picked up the habit of writing priors as testable statements from the scientific framing Michael Watkins uses in The First 90 Days, then made it mechanical because willpower is not a system.

Hold conclusions for 30 days

One rule governs the whole month: no broadcast conclusions before day 30. You are allowed observations and questions. You are not allowed a thesis.

The reason is mechanical, not modest. The moment you share a thesis, every subsequent interview becomes a referendum on it. People who agree amplify. People who disagree go quiet or go political. Your information supply curates itself to match your stated belief, and the ledger fills with echo. I have watched new executives leak a thesis in week two and spend the next six weeks collecting confirmation they mistook for evidence.

When someone pushes you for a verdict, and they will, the script is: "I have a list of claims and I am still checking them. Ask me at day 45 and I will show you the list." Saying "I will show you the list" turns the deferral into a commitment, which goes straight into your trust ledger.

Failure modes

The tour becomes therapy. Some interviews turn into an hour of venting. Useful for rapport, useless for the ledger. Let the venting run five minutes, then redirect: "That sounds real. Help me make it checkable. What would I look at to confirm it?" If they cannot answer, log the emotion as context, not as a claim.

The leaky thesis. Covered above, but it deserves its own line because it is the most common failure. It usually leaks through a "great question" reaction or a leading follow-up, not a declaration. Keep your follow-ups symmetric: "what would someone who disagrees with that say?"

Anchoring on the most articulate person. Every org has one person whose model of the company is so crisp and well-delivered that it colonizes yours. Flag their claims in the ledger and deliberately hunt for disconfirmation. Crisp is not the same as correct, and the most articulate person is often articulate because they have been polishing one story for years.

The ledger nobody rereads. A write-only ledger is a diary. The Friday synthesis is what makes it an instrument. If you skip two Fridays, you are back on the tour.

A worked example

Generic and sanitized, but I have watched this exact shape more than once. A new CPO logs the claim "we lose enterprise deals on missing SSO" from three sales interviews. Old tour behavior: SSO goes on the roadmap by day 40. Ledger behavior: confidence stays low because the source is one function, evidence needed says "read the last ten loss reports." The loss reports show SSO mentioned twice, and in both cases the deal was already dead on price. The contradicting claim, from a quiet CS manager in interview twenty-two, was "deals stall because security review takes six weeks and nobody owns it." That claim checked out. The fix was a process owner, not a feature, and it cost two weeks instead of two quarters.

That is the whole argument for the ledger in one anecdote. The tour would have shipped SSO.

Pick one thing this week

If you are in your first month: take your existing interview notes, however messy, and convert the ten strongest assertions into ledger rows. Claim, source, confidence, contradicts, evidence needed. The conversion itself will show you how much of what you have collected is checkable and how much is vibe.

If your first month is behind you: ask the next five people you meet "what should I kill?" and log the answers. It is never too late to start the file.

The ledger template, the nine questions as a printable script, and the Friday synthesis checklist are in the CPO First-90 Kit(coming Jun 10). The full 90-day arc is in The CPO 30/60/90.

Sources: Michael Watkins, The First 90 Days for the transition frame this post rebuilds, Teresa Torres, Continuous Discovery Habits for the interviewing-as-evidence discipline, Marty Cagan / SVPG on product leadership transitions.

Further reading

• CPO Days 31 to 60: Instrument the Org Before You Steer It(coming Jun 15)
• CPO Days 61 to 90: The Kill List, the Bets, and the Readout(coming Jun 24)
• The Interview Guide