How it works

Describe what's going wrong in plain words, and paste any numbers, tickets or observations you have.
Pick the type: product or UX, business, project, or process improvement. Click Write my problem statement.
Share the statement with your team before anyone proposes a solution. Fill in the [evidence needed] items first if there are any.

Examples

Sample results for the inputs shown, so you can see the format and quality before you try it.

Example 1: input

What's going wrong? New customers sign up for our scheduling app but a lot of them never add their team. Support says setup is confusing.

Type: Product or UX problem

Data or evidence you have: 38% of new accounts never invite a second user in their first 14 days. 11 of the last 20 support tickets from new accounts were about setup. Accounts with 2+ users are the ones that convert to paid.

Example 1: output

Problem statement: Office managers who sign up for our scheduling app often never get their team into it: 38% of new accounts don't invite a second user in their first 14 days. Accounts with two or more users are the ones that convert to paid, so every account stuck at one user is likely lost revenue. Support tickets suggest setup is confusing, but we haven't confirmed where people get stuck. Who is affected: people setting up a new account for their team, usually an office manager or owner. What's happening: 38% of new accounts never invite a second user within 14 days. Evidence: 38% single-user rate at day 14; 11 of the last 20 new-account support tickets were about setup. Impact: Accounts with 2+ users are the ones that convert to paid, so single-user accounts are unlikely to pay. [impact: multiply stuck accounts per month by your paid conversion rate] Success looks like: the share of new accounts with no second user by day 14 falls from 38% to 25% (suggested). Not in scope: why multi-user accounts churn later. How might we: How might we help a new account owner get their first teammate in during their first session?

Example 2: input

What's going wrong? Our month-end close takes too long. Finance team is 4 people, lots of manual reconciliation between Stripe and QuickBooks.

Type: Process improvement problem

Data or evidence you have: Close takes 9 business days on average over the last 6 months. About 60% of the time goes to reconciling payouts. Leadership wants numbers by day 5.

Example 2: output

Problem statement: Our 4-person finance team takes an average of 9 business days to close the books each month, against leadership's goal of 5. About 60% of that time goes to manually reconciling Stripe payouts with QuickBooks, which delays the monthly numbers leadership uses to make decisions. Who is affected: the 4-person finance team, and the leaders waiting on monthly numbers. What's happening: month-end close averages 9 business days over the last 6 months. Evidence: 9-day average close; roughly 60% of close time spent reconciling Stripe payouts to QuickBooks. Impact: numbers arrive 4 days later than leadership wants each month. [impact: add team hours spent on reconciliation per month] Success looks like: average close time falls from 9 business days to 5, measured over 3 consecutive months. Not in scope: changes to the chart of accounts or the reporting package itself.

Example 3: input

What's going wrong? We keep missing project deadlines on client website builds. Clients are getting frustrated.

Type: Project problem (for a proposal or charter)

Example 3: output

Problem statement: Our client website projects are regularly delivered after the agreed deadline, and clients are becoming frustrated. We don't yet know how often it happens, by how much, or at which stage projects slip, so we can't fix the cause with confidence. Who is affected: clients waiting on their websites, and the project teams delivering them. What's happening: website builds miss their agreed delivery dates. Evidence: [evidence needed: for the last 10 projects, the agreed date, the actual date, and the stage where each one slipped] Impact: frustrated clients. [impact: count projects with complaints, discounts or lost follow-on work linked to delays] Success looks like: [target] share of projects delivered by the agreed date, measured over the next [number] projects. Not in scope: pricing and how projects are sold.

Tips for better results

Paste your numbers. "38% never invite a second user" makes the statement specific; without data you'll get [evidence needed] notes telling you what to measure.
Leave your favorite solution out of the notes, or expect it to be moved aside. A problem statement that names the fix skips the most useful debate.
Name the person affected. "Office managers setting up an account" leads to better ideas than "users".
Agree on the Success looks like line before starting work. It's how you'll know later whether the fix worked.

FAQ

Is this problem statement generator free?

Yes. Enter your email once to use it (you'll also get Something Big, our free weekly AI newsletter, and you can unsubscribe anytime). There's no account and no credit card.

What makes a good problem statement?

It names who is affected, what is happening in observable terms, the evidence, the impact, and what success would look like, without proposing a solution. That's the structure this tool writes.

Can I use it for Lean or Six Sigma projects?

Yes. Choose Process improvement problem and it describes the current process, where it breaks and the measurable waste, with a baseline and target for your charter.

What if I don't have data yet?

It still writes the statement, and replaces missing numbers with [evidence needed] notes that say exactly what to measure and where to find it. It never invents figures.

More free AI tools

Related prompts

Prefer to use ChatGPT or Claude directly? These free prompts do similar jobs.