I Like I Wish What If

I Like I Wish What If is a critique framework borrowed from IDEO in the early 2000s and sharpened for design teams that need to review early stage work without destroying momentum. The process is dead simple. The designer posts the work and states their goal in two sentences or less. The team then responds in three dedicated rounds. In I Like they call out specific elements that are working and should stay in future versions. In I Wish they name aspects that fall short and need improvement. In What If they propose speculative changes or entirely new directions to explore. This structure replaces the usual unstructured mess with a script that keeps everyone on the same page. It forces positive feedback first which counters the natural human tendency to lead with criticism. The wish language softens feedback so it feels like collaboration instead of judgment. The what if round opens the door to creative leaps that rigid feedback formats kill. This framework integrates perfectly with the three layer feedback model. I Like surfaces observations about what succeeds. I Wish captures interpretations of where the work misses the mark. What If delivers suggestions in an open ended form that respects the designer as owner of the work. Teams at companies like Figma and Stripe have used versions of it to cut through the theater that plagues most design reviews. Instead of seniors monologuing about button colors for 20 minutes the conversation stays focused on moving the work forward. It directly solves the five common failures. Designers ask for specific feedback instead of presenting. Reviewers stop competing because the format gives everyone equal airtime. Personal taste gets exposed when it cannot fit into a clear I Wish tied to user or business goals. The required note taking captures everything for later review. And because it works best early it prevents the too late problem that wastes everyone's time. What it is not is a complete critique system on its own. It does not replace the need for a facilitator who keeps the conversation on rails and enforces the rules around one conversation at a time and specificity. It is not appropriate for final decision making or high stakes pre ship reviews where the Decision Stack framework delivers better results. This is not a method for giving prescriptive orders. Any reviewer who says I wish you would just make the button blue misses the point entirely and reveals they skipped the observation layer. The framework assumes the team already follows the rule of pairing every suggestion with the observation that supports it. Without that foundation the wishes become demands and the what ifs become noise. It also is not a shield for bad work. Just because someone says I like the overall direction does not mean the designer should ignore deeper structural problems that surface in the other rounds. Consider how the framework played out for the Intercom team in 2022. They were redesigning their inbox interface for better focus modes. The early mocks went into an I Like I Wish What If session. Reviewers liked the new keyboard shortcuts that surfaced at the right moment and made power users faster. They wished the distinction between read and unread messages was stronger in dark mode because it broke scanning patterns. The what if round generated ideas like what if we used subtle animation to draw the eye or what if we borrowed patterns from email clients like Superhuman for priority threading. The designer synthesized those comments into two new directions. One became the core of their new focus feature that users praised in the release notes. The session took 15 minutes. The notes from the meeting guided the next iteration without any need to rehash the same points two weeks later. A second example happened at Shopify in 2024 during their checkout optimization project. The design lead presented three early variants in a FigJam board tied to their goal of reducing cart abandonment. The I Like comments praised the simplified form labels that reduced cognitive load on first time users. Several team members wished for better error handling visibility before users hit submit because current patterns hid failures until too late. What if suggestions included what if we used inline validation with illustrations or what if we moved error messages to a dedicated banner at the top that persists on scroll. Those inputs helped the team identify a hybrid approach that combined the best elements. Post launch data showed a 19 percent lift in completion rates. The framework prevented the team from jumping straight to solutions and forced them to articulate what was working first so iterations built on strengths instead of starting from zero. Use I Like I Wish What If during the direction setting phase of any project. The main article recommends it specifically for early reviews where the goal is to validate the problem solution fit and generate options before anything gets polished. It works beautifully in remote settings where you post the work in Figma or Miro with clear prompts for each category and a 48 hour deadline for responses. Async versions often yield more thoughtful input because people respond on their own schedule instead of performing for the room. Pair it with the rule that the designer scopes the feedback upfront so comments stay on hierarchy or flow instead of veering into locked copy or brand colors. It shines with cross functional teams because the simple language does not require design training to participate and PMs or engineers can contribute what ifs that expose product blind spots early. Avoid this framework once the work reaches mid fidelity or later. At that point the questions get more specific and the Goal Rationale Feedback framework or the full Decision Stack produce tighter outcomes on what ships. Do not use it with teams that have not yet adopted the basic critique rules around specificity and capture because the structure alone cannot fix a broken culture where people use the meeting to show off or settle scores. It also falls flat for very small changes or bug fixes where a quick three layer feedback pass works faster. If the team consistently turns what ifs into arguments about unrelated topics park those items immediately and switch frameworks before the whole exercise loses credibility. I Like I Wish What If lowers the temperature in critique so teams can actually hear each other and improve the work instead of defending their egos.