Example · Product Owner
Story writing is going to tooling. Claim the decisions behind it.
25 of your 40 hours still need you
- 20 stay yours
- 5 faster with AI
- 15 tools could take
A typical week at the middle pace. Your plan uses your own tasks.
Is product owner a dying role?
By our task-by-task count, 25 of a typical 40-hour week as a product owner still need a person. Tools could already take about 15 hours. A model already does the part your title is named for, and it is 37% of your week. That includes stories from a decided feature, acceptance criteria, backlog tidying, release notes and the review deck.
Read more
If you have watched Jira's AI write an epic's worth of stories, you have seen it. The acceptance criteria come out cleaner than yours. One or two are wrong about how your product works, and nobody catches it until a developer builds the wrong thing. That gap is what this report is about.
The half that holds is the part nobody wrote a job posting for. It means knowing which customer is the renewal and saying "that is not done" to an engineer in front of the squad. It also means telling a stakeholder with budget that the request is not in the sprint, and ordering the backlog so the squad builds the right thing first. The Scrum Guide puts all of that under one word, accountable. Many companies are deciding whether that accountability lives with the Product Owner or with a Product Manager who never writes a story, and they have been merging the two seats since before any agent could draft a story. That is the risk to name first. The plan is about making the half that holds yours in writing, before the decision is made for you.
The hard truth
The backlog can be delegated, but the value cannot.
Scrum itself lets a Product Owner hand backlog writing to others. What you cannot hand off is accountability for the product's value.
Sources (1)
- The Product Owner may delegate backlog work to others but remains accountable for maximizing the product's value. The Scrum Guide (Schwaber and Sutherland), 2020-11
What stays yours
Knowing which customer actually matters
The renewal calendar, the escalation history and the fact that one customer's executive sits on the advisory board are not in Jira. A model can rank requests by frequency, but it cannot know that the quiet customer is the one that pays.
Saying "not done" at sprint review
The Scrum Guide gives the Product Owner the acceptance decision. Saying it to a person, in a room, with the squad watching, is what makes the next sprint cleaner. A tool can check the criteria, but it cannot be the person who answers for accepting the work.
Telling a stakeholder with budget "not this sprint"
The Scrum Guide says those wanting to change the backlog "can do so by trying to convince the Product Owner." So someone has to be the person who can be convinced, and can decline.
Ordering the backlog
Ordering wrong looks fine for a quarter. The cost shows up as a missed renewal or a feature nobody uses, months later. A model can propose an order from the data, but it cannot carry the cost of being wrong.
Catching the generated criterion that is wrong about how your product works
A well-formatted wrong story is more expensive than a messy right one, because it gets built. The person who can read a generated criterion and say "that is not how approvals work here" is the constraint on every story-generation tool.
What's next after product owner?
Plan A is your best move inside the work you do now. Most sessions in the 13-week plan build toward it.
- A
Own one outcome, one customer set and the "no", in writing
Why it fits
Why you. If you already decide what is done, already know which customer matters and already take the executive's call, you have standing nobody else at your level has. The gap between what you do and what your title says is the smallest of the three moves.
The gap. Most teams have no written record of stakeholder requests. Accepting work is a click in Jira with no log, and no squad has an outcome metric with a person's name on it.
The trade-off. You become the person blamed, in writing, when the executive escalates and when the metric misses. The hard moments stay, but you have more authority on paper. You will also write fewer stories for a quarter, and someone will count.
Other moves that fit
These are also good bets in this field. Read them if Plan A doesn't suit you.
Own the quality of machine-written stories
Why it fits
Why you. Finding the generated criteria that are wrong about how your product works is a skill with no name yet, and story generation is going to every squad.
The gap. There is no written standard for what an acceptance criterion must say and no guidelines for the AI tool. Nobody has measured how often generated stories are wrong.
The trade-off. You read more than you write, and some weeks it feels like proofreading for a machine. It is also the task most likely to be handed to a product operations seat in a reorg. Without the outcome move, this one makes you the best-qualified person for the seat that edits.
The portable hedge: domain product owner in your vertical
Why it fits
Why you. Knowing which account is the renewal and how a tricky workflow works is worth money at every software company in your vertical. Domain product work is the corner of this occupation where hiring depends on context more than story-writing.
The gap. You have no public work sample yet, and you don't know what the market pays for the title your company will use.
The trade-off. Leaving resets the asset this report says protects you. Your first year anywhere new is your most exposed: context starts at zero and story-writing is all anyone can measure.
What's a plan B for product owners?
Plan B is a job AI is creating that these skills already fit. One session a week in the plan trains you for it, so it is ready if your job shrinks or you want a change.
BPlan B, picked for this role
AI product owner
Owning the backlog and saying no in writing is the job, so taking the AI feature on your current team is the usual way in.
Other options
AI product manager
Tying one outcome to one customer set is product management thinking, and leading an AI feature end to end builds the case.
AI evaluation specialist
Writing acceptance criteria that define done prepares you to define what a good AI answer looks like and score against it.
In your own plan you can switch to any of the 47 roles.
The first two weeks
The first 2 of 13 weeks. Three sessions a week of 45, 60 and 90 minutes, plus one 40-minute Plan B session. Weeks 4, 8 and 12 line up with a sprint review every two weeks, so move them to match your team's schedule. By week 13 you'll have a written intake record for stakeholder requests, an acceptance log with numbers next to your name, and your product lead's answer on making those decisions yours.
- Week 1: Measure your week
What you do
- ALog how you spend your time for one week. Sort each hour into the tasks listed in this report.
- APull numbers for your last three sprints from Jira, your team's ticket tool. For each sprint, count stories written, stories rejected at review, and bugs that reached customers.
- AOpen an epic, a large feature split into stories, that the AI wrote, or have it generate one. Sort every story into right, well formatted but wrong, or missing something the ticket left out.
- BTake Anthropic's free AI Fluency: Framework & Foundations course. Then ask Claude, Anthropic's AI assistant, to draft five stories for a made-up app and mark which criteria are wrong.
- Week 2: Learn which decisions are yours
What you do
- AAsk your product lead in writing which decisions are yours. Name three: accepting work, ordering the sprint, and declining a stakeholder request.
- ARead the Banking Dive article on Capital One and the Product Owner section of the Scrum Guide, the official Scrum rulebook. Write one page on which tasks the guide gives you and which your product lead holds today.
- AFind out who at your company will decide whether the Product Owner and Product Manager roles merge, and when.
- BTake PMI's course Talking to AI: Prompt Engineering for Project Managers. PMI is the Project Management Institute. Build a chain of prompts that turns a made-up request into a story with criteria.
Three courses from the plan
Checked 2026-10-11.
- Professional Scrum Product Owner I (PSPO I)Scrum.org · $200 per attempt
Only if a posting demands it and you do not already hold CSPO. Its only advantage is that it costs less than renewing CSPO for life. Take the free Product Owner Open assessment first.
- Professional Scrum Product Owner II (PSPO II)Scrum.org · $250 per attempt
No. It certifies your reading of the document, and the people merging your seat do not look at certificates.
- Professional Scrum Product Owner III (PSPO III)Scrum.org · $500 per attempt
No. Not this year.
Risks beyond AI: highSeat-merge risk
This has nothing to do with AI. Companies have been folding Product Owner seats into Product Manager or engineering-lead seats since before today's tools existed. Capital One eliminated about 1,100 roles in its "agile" job family in January 2023 and said the duties would move into existing engineering and product manager roles. If your company is saying the POs will become PMs, ask which half of the job the merged seat keeps. We score this separately because blaming AI for it would get the remedy wrong.
Where this could be wrong
- Developer questions may be more exposed than we scored. We assumed the hard half is what the ticket left out. If the squad points the AI at your wiki, call recordings and chat history and the questions to you drop by half in Week 7, AI can answer more of those questions and the story quality move becomes urgent.
- We assumed acceptance is yours informally. If your product lead or the engineering lead already accepts and you merely transition the ticket, you carry less responsibility for acceptance, the outcome move is "join the decision" rather than "own it," and the Week 4 log is harder to start.
- The merged seat may already be decided. If the reorg is written and the product operations seat is yours by name, the Week 12 ask is late and the Week 13 application comes first.
- The headcount figures are not a clean series. The 28% is US tech PM headcount from a private data vendor summarized by a trade publication, and the UK counts are small. The direction matches what many companies are saying out loud; the size is not a measurement of your market.
Also matched: product owner, po, agile product owner. Research checked 2026-10-11. Page updated .