先把 punchline 抛出来：Figma 演示团队怎么用产品评审赢到信任，而不只是一盏绿灯

One of the speakers opens with a photo from their Uber days, taken during a first product review with Travis Kalanick. The review collapsed into slide-flipping: the CEO took over and effectively used the PM as a remote control, eventually saying he would just read the deck himself. The CEO probably got what he needed, but the PM lost control of the narrative and could not direct attention to the areas that needed scrutiny. In talking to many PMs — and sitting on the reviewer side — the speaker found that a lot of product reviews feel broken in exactly this way.

Mika and Yuki argue the root cause is that PMs are given advice about product reviews that sounds right in theory and is often wrong in practice. The most common version: a review exists to make a decision between radical option A and radical option B, and it is the only place you can get a green light. They disagree, arguing that the most important outcome of a product review is not the decision at all but building trust and winning over key stakeholders. The talk is organized around the things PMs have been coached to do that backfire, and the taboos that actually lead to success.

At Figma, product reviews usually take the form of slide decks in a 25- to 30-minute slot, so that format dominates the examples. The speakers stress that the takeaways apply to any forum — a deck walkthrough, a hallway huddle over a laptop, or an informal conversation. The goal is to change how PMs think about the room, not just how they build slides.

A few minutes into a typical review, the PM is still setting context. In Mika's first product review at Figma, she started screen-sharing and immediately got a ping from Dylan asking, "does anyone have the link?" While she talked about why the team was excited about voting in FigJam, she watched Dylan scroll all the way to the end of the deck, ignoring the narration. Over time she learned to translate: "does anyone have the link" was exec-speak for "what is the actual recommendation?"

Later in the meeting the PM often pauses to check alignment: "I've been talking a lot — are we aligned on the problems, the goals, the principles?" That is usually met with an awkward silence. Reviewers may be tuned out because the framing did not seem important, may be multitasking in Slack, or may simply be unable to agree yet. The silent question is often "is this a trap?" — because agreeing to framing without knowing the recommendation risks endorsing the wrong one. Reviewers want to cheat and read ahead before conceding alignment.

The awkwardness comes from the standard template PMs are taught: agenda, background, problem, goals, principles, then solutions and the recommendation. It looks sensible, but it buries the spicy proposal in the final minutes, after everyone has sat through context they are tuning out or only mildly agreeing with. The good reactions — positive or negative — arrive far too late. And when the framing, problems, or goals are themselves controversial, that is impossible to detect before the audience knows the recommendation.

Even though it violates storytelling instincts, in a review the PM should start with the punchline. The point is not suspense or a gradual build — it is to reach the part that feels controversial quickly, because that is what makes everyone pay attention and engage. If starting with the spicy content produces a violent reaction, that is a good thing: the room is now actually in the conversation.

A simple gut check is whether the interesting content arrives within the first 20% of the deck, or the first five minutes of the review. In Figma Slides' own Vision deck, the valuable material — key concepts, product thinking, differentiation — was concentrated in a small number of slides that the team tracked deliberately. Many decks are far more top-heavy with background and context than PMs realize, which sets the meeting up for failure before it starts.

There are subtler levers too, starting with title cards. A title that says "audience," or more dramatically "who is our audience?", forces people to guess. A title card with a point of view cuts to the chase, so people pay attention up front and a distracted reader can scan and immediately re-engage in the discussion.

Sometimes the PM genuinely does not know which part is spicy, and therefore where to spend discussion time. Figma's ritual for this is an "alignment widget" sent out beforehand that solicits everyone's opinions and reveals where misalignment is greatest, so the meeting can be spent on the areas of real disconnect.

When assembling a deck, PMs try to cover their bases and pack every section with new information. The speakers' counter-argument is that if you say everything, what everyone hears is nothing. Instead of covering all bases, PMs should be coached to make a meme — something that goes internally viral and lodges in the company's memory.

The original Figma Slides Vision deck carried two memes. The first was the name: internally the product was called "Flides," deliberately chosen over generic options like Figma Slides, Figma Decks, or Presentations. For an entirely net-new product to earn a place on the roadmap through bottom-up energy, anyone at the company — a tenured engineer or a first-day hire — needed to hear the term and immediately want to learn more. It spread: the animation sync became the Flides animation sync, the layout workstream became the Flides workstream, Friday Zooms started with "happy Flides day," and Mika's mom became irrationally attached to the name and insisted it not be abandoned.

When the product went to market in June, the quirky internal name paid off externally: Twitter blew up with thousands of impressions from people excited to try the product. A fairly normal product launch carried an almost emotional, irrational attachment that people had formed to a word.

The second meme was a single stat: over 3.5 million people had made slides in Figma design over the previous year. That number proved the investment rested on a real, currently unsolved pain point rather than an interesting product idea — millions of people were jumping through acrobatics to hack a tool that was not built for slides. It gave the team internal traction and made a bet in the space feel obvious. At the Figma Slides announcement, the same stat got thousands of designers in the audience to raise their hands, and nearly everyone was shocked at how pervasive the workaround was around them. The coaching point: don't obsess over the five or ten stats people should remember — find the one that tips an executive over the edge.

Seventeen minutes into a review, someone asks an unexpected question despite the PM's request to hold questions to the end. The classic response — "great question, I'll circle back" — quietly degrades trust. Hearing "great question" can signal that this was the first time the PM considered something basic, and promising to follow up later feels slow to an executive who wants the answer now; waiting a week feels extremely frustrating.

The alternative is to be unafraid to share your gut in the room. PMs often withhold assumptions because they lack perfect information, and the cost is discovering a week later that everyone had very different assumptions — a week burned, back to square one. Sharing a rough, imperfect point of view surfaces whether people are on the same page, and lets the team get to the bottom of the misalignment immediately.

One of the speakers recalls that at Uber, then-CPO Jeff Holden called this "cutting out the wait space": he saw the executive's job as removing the arbitrary amounts of time people spend waiting on others to make progress. Since PMs often want to run an experiment, study, or sprint first, Holden used a trick — framing it as a thought experiment, fast-forwarding to the day the experiment is over and asking what the team expected to see.

That framing lowers the stakes, because everyone knows they are guessing, and it creates a safe space to share rough thinking. The practical instruction to PMs: instead of taking an action item to circle back, say what you think you will find — it saves everyone time and builds alignment far faster.

With only a few minutes left, PMs tend to talk only about the positive: why this is an exciting investment and why it beats the alternatives. The executive in the room is usually thinking about the downsides, and in exec-speak is wondering whether the PM has thought deeply — or is hiding something. At that point trust breaks. A big bold pitch matters, but treating your own idea with skepticism matters just as much.

The official Figma Slides Vision pitch in February was preceded by two earlier review-like forums. The first was Maker Week, Figma's internal hackathon: the team recruited designers and engineers around the New York office, built a prototype in one week, and structured the demo like a product review for the entire company. The chat blew up, including people like Dylan who had previously been skeptical — and when the team asked for a green light, the answer was still no. A hype Zoom pitch, it turned out, does not secure buy-in.

The next attempt moved into a formal product review: competitive landscape, why and how Figma should win in storytelling, and the plan to combine Figma's visual control with FigJam's interactive storytelling mechanisms. The competitive matrix came out with Flides holding all the green boxes and every other player marked down — to the PM it looked like an obvious investment case. In reality it was stacking the cards, and the answer to "do we have a green light" was still essentially no.

The official Vision pitch was both a bold pitch and an exercise in skepticism. The team spent weeks talking to users to understand pain points, then structured the deck so that every pain point was immediately followed by a solution and a proof point testing the hypothesis. Where the team was genuinely unsure — notably how to build a tool where designers and non-designers each keep their preferred toolkit — the deck laid out the competing variations honestly rather than pretending certainty. Finally, they preempted the spice: identifying the hard questions an executive might have and proving the homework had been done to answer them.

The speakers close by reframing why all this matters: stakeholders are human beings with short attention spans who are easily distracted, context-switching, and emotionally swayed by how a review makes them feel. The best PMs know this and use it. Making decisions still matters, but winning trust is what allows decisions to be made faster over the long run.

The biggest myth of all is that the product review is a monumental, one-shot forum. Despite the positive feedback and immediate internal traction after the Vision pitch, Mika was fixated on whether she had received a "formal green light" and messaged Yuki right after the presentation to ask. The reality is that what happens outside the review — the days, weeks, and months spent building with the team — matters far more. The real question is whether a review buys enough trust to do the thing that matters most: build.

Yuki notes that at the start of the year Figma had three company priorities and Flides was not one of them. Leadership was excited about the work and the vision but unsure whether it was the right time to launch a new product, so Mika focused on making it real — moving from a cool idea a few people were exploring to a product the company would put itself behind. She found an opening in the sales kickoff, the company's biggest internal event of the year, and proposed that the keynote be presented in Flides. The idea of putting an unshipped product on stage was not to leadership's liking, and they pushed through anyway, with a barely ready product and plenty of bugs in the run-up.

The day of the keynote went perfectly. Yuki presented alongside CTO Chris — apparently always in blue sweaters on stage — and the presentation ended with a reveal that the deck the audience had been watching was made in Flides all along. Five months later, at Config, Figma Slides launched as the number-four hero moment of the event and audiences loved it. The lesson: the initial product review bought the trust to explore the idea, but getting the product usable in front of a live audience bought the trust to bring it to market. The team also open-sourced the deck itself on Figma Community so others can remix the animations and structure for their own reviews.