Most organizations I work with treat Working Backwards as a one-shot artifact. They run the workshop, write the Future Press Release, file it, and go back to the roadmap. Two months later, no one is sure whether the PR ever shaped a decision.
Working Backwards should not be seen as just a way to create a polished Future Press Release. If everyone agrees it's a great document but nothing changes, I consider it a failure.
The Future PR describes a beautiful tomorrow with delighted customers. But if no one has named today with the same precision, the gap between the two stays abstract, and abstract gaps rarely move a team.
I want to share a four-step workflow I now use with leaders to fix this. Working Backwards, then Today Statements, then a PIEES overlay, then a rapid prototype. The first two force clarity. The third expands the idea set. The fourth collapses the time from insight to evidence. Done well, the whole exercise runs inside a single week.
Step 1: The five Working Backwards questions
Amazon's Working Backwards method is built on five customer questions:
- Who is the customer?
- What is the customer problem or opportunity?
- What is the most important customer benefit?
- How do you know what the customer needs or wants?
- What does the customer experience look like?
I covered this method in Issue #23. It works because it forbids a team from writing the solution before they've named the customer. But even when the workshop has gone extremely well and led to a crisp PRFAQ document, there is no guarantee it will create value anytime soon.
An important part of the Working Backwards problem-definition phase is aligning on a Today Statement.
Step 2: The Today Statement
A Today Statement is a single structured sentence that captures the answers to Working Backwards questions 1, 2, and 3 (customer, problem, most important benefit), written in the present tense about today.
The template I teach is:
Today [customers]
Have to [customer problem]
When [situation when the problem occurs]
[Customers] need [most important benefit]
Here's the example I use in my workshops:
Today Swedish residents, who have a family with young children and manage their family's affairs, have to remember different passwords and go through multi-step verification processes, whenever they need to use digital services at banks or public services or education providers. They need a simple way to use digital services which require secure identity verification.
The Today Statement locks the answers to questions 1 to 3 before anyone imagines the future. It makes the gap between today and tomorrow concrete enough to act on. The ideas that follow, and the Future Press Release that they will lead to, become the destination. The Today Statement becomes the pressure test.
Step 3: Overlay PIEES on the gap
Once the gap between today and the need is named, the question is how to close it.
I use PIEES as one way to surface ideas to close that gap. Personalization, Interaction, Emotion, Experiences, Stories. I introduced the framework in Issue #25 and extended it into data flywheels in Issue #26, so I won't re-explain it here. For each PIEES lever, ask: if we closed the gap through this lens, what would the customer experience look like? You'll end the hour with 20 to 30 candidate ideas instead of one obvious chatbot.
Pick the two or three with the highest value-to-effort ratio. Those go to Step 4.
Step 4: Prototype during the same week
The cost of being wrong has collapsed. In 2026, a small set of tools makes this step real.
With tools such as Base44, Lovable, or Claude Code, a team can test three PIEES overlays before the end of a single meeting, and walk into the next one with working apps rather than slidedecks. The prototype does not have to be polished. It has to be real enough to put in front of a customer or a stakeholder and watch what they do with it.
Your week, Monday to Friday
If you want to run this yourself, here's the shape of the week.
- Monday — answer the five Working Backwards questions.
- Tuesday — write the Today Statement using the four-part template.
- Wednesday — overlay PIEES on the gap and pick two or three candidates to build. If you have extra time, use an AI tool to draft a PRFAQ document.
- Thursday — prototype day in whichever tool fits what you're trying to learn.
- Friday — the team reviews what actually got built and decides what to take further.
The real power of this workflow is in the sequencing. Working Backwards without a Today Statement produces beautiful PRs that nobody acts on. Today Statements without PIEES default straight back to the productivity lens. PIEES without a prototype stays theoretical. Run all four together, and you compress a quarter of discovery work into a week.
How this connects to the rest of this issue
The leadership section of this issue uses scenario planning to surface which futures you should bet on at all. The terminology section explores Personal Context Management — the curated context that makes your AI work feel like yours. Scenarios decide what to bet on. PRFAQs and Today Statements name the bet precisely. Prototypes turn the bet into evidence. Personal context makes each step better than the last.
Your action step
Block a single week on your team's calendar. Pick one customer problem worth being right about. Run Monday-to-Friday in the order above. At the end of Friday, the success test is not "did we ship." It is "do we know more than we did Monday, in a form we can defend in front of a customer?"
If you need a tool to be your sidekick, I've built a Base44 tool called OutcomeHack that walks teams through the five Working Backwards questions and stress-tests the answers.
If you want me to run this week as a workshop with your leadership team, I'd love to help.