Inefficient decision making costs a typical Fortune 500 company around 530,000 days of managers’ time every year, roughly $250 million in wages, according to a McKinsey survey of more than 1,200 global business leaders. The same research found executives spend close to 40 percent of their time making decisions, and 61 percent of them believe at least half of that time is wasted. Only one in five say their organization is actually good at it.
Sit with those numbers for a second, because they explain something most product managers misread. When you walk into a leadership review with three tidy options and no point of view, you are not being neutral or thorough. You are handing an already overloaded decision-maker one more decision to make, and doing it in the exact format that produces the waste McKinsey measured. The room does not experience your even-handed slide as rigor. It experiences it as you making them do your job.
I have sat on both sides of that table. Twenty-plus years in IT operations, and later fractional COO work, put me in a lot of rooms where someone presented “the options” and then waited. The presentations that moved fast had one thing in common, and it was never the quality of the deck.
The three-option slide is where decisions go to die
There is a specific move I want to name, because almost every PM does it at some point and almost nobody gets told to stop. You research a problem, you find three or four plausible paths, you build a slide that lays them out side by side with pros and cons, and you present it. Clean. Balanced. Defensible.
And then nothing happens.
The VP asks a clarifying question. Someone from another team raises a concern about option B. The head of sales quietly lobbies for the one that helps this quarter. Forty minutes later the meeting ends with “let’s take this offline” or “can you come back with more detail,” which is corporate for you did not actually help us decide anything.
Here is what went wrong. You did the hard 80 percent, the research, the trade-off analysis, the framing, and then you stopped right before the part only you were positioned to do. You know this problem better than anyone in that room. You talked to the customers. You saw the data. You understand the constraints. The approver did not do any of that, and now you are asking them to arrive at the right answer in twenty minutes using a summary you wrote. That is a worse process than the one that produced the McKinsey number, not a better one.
Jason Knight, who writes the One Knight in Product newsletter, calls the neutral version of this “weather reporting,” and the phrase has stuck with me. A weather reporter tells you it might rain, might not, here are the percentages, good luck out there. A weather reporter has no stake in whether you bring an umbrella. When a PM presents options without a recommendation, Knight argues, they are giving away the one thing that makes them valuable: the chance to influence the decision. His line for it is blunt and correct. You touch it, you own it.
What “make a recommendation” actually means
Making a recommendation is not the same as hiding the alternatives or pretending you did less analysis than you did. The alternatives still belong in the room. The difference is where your weight lands.
A recommendation has four parts, and if any one is missing the room feels it:
First, the pick. One option, stated plainly, in the first sentence, not buried on slide nine. “I recommend we build the lightweight version for enterprise accounts and hold the self-serve flow until Q1.” If a stranger read only your opening line, they should know exactly what you are asking for.
Second, the reasoning. People will not commit to a decision whose logic they cannot see. Bring them into it. Show the data, the customer evidence, the constraint that made the call. The goal is not to win an argument; it is to make your thinking transparent enough that a smart person could either agree or find the flaw.
Third, the alternatives you rejected, and why. This is the part that separates a recommendation from an opinion. Say what you considered and what specifically made you set it aside. Naming the option you are not recommending, and being honest about its appeal, does more to build trust than any amount of confidence. It signals you actually looked instead of defaulting to your favorite.
Fourth, your confidence level. Not every call deserves the same conviction, and pretending it does is how PMs lose credibility over time. Itamar Gilad’s Confidence Meter is a useful discipline here: weight how strongly you argue by the strength of your evidence and the size of the bet. A reversible, cheap experiment gets a fast, loosely held recommendation. A one-way decision you cannot easily undo gets more evidence and more caution before you plant a flag.
That last distinction is not mine. It is Amazon’s, and it is worth stealing.
Borrow the door test before you decide how hard to push
In his 2016 letter to shareholders, Jeff Bezos split decisions into two kinds. Type 1 decisions are one-way doors: hard or impossible to reverse, so they deserve deliberation and a heavier evidentiary bar. Type 2 decisions are two-way doors: you walk through, and if you do not like what you find, you walk back out. Most decisions, Bezos argued, are two-way doors that organizations mistakenly treat as one-way doors, and that mistake is a major source of the slowness McKinsey later put a price tag on.
For a PM, the door test changes how you present. If you are recommending a two-way-door move, say so, and use it to unstick the room: “This is reversible. If the enterprise version underperforms in six weeks, we roll it back and we have lost six weeks, not the roadmap.” That single sentence removes most of the fear that makes stakeholders hedge. You are not asking them to bet the company. You are asking them to try something you can undo.
And when consensus genuinely will not come, Bezos gave everyone permission for the other half of the move: disagree and commit. His example was greenlighting an Amazon Studios show he personally doubted, telling the team, in effect, I disagree and I commit, and I hope it becomes the most-watched thing we have ever made. A PM can offer that same deal upward. “I know you are not sold on this. Will you commit to it with me for one cycle and let the results settle it?” That is a far stronger position than another options slide, because it moves the room from debating forever to deciding now.
Make the roles in the room explicit
Part of why option slides stall is that nobody knows who is actually deciding. The most useful fix I have seen is also one of the oldest: the DACI model, developed at Intuit back in the 1980s as a variant of the RACI matrix. It assigns four roles for any decision. The Driver moves it along, usually the PM. The Approver makes the final call, singular, one person. Contributors supply expertise and get consulted. The Informed are told the outcome.
The point is not the acronym. The point is that a lot of decision waste comes from ambiguity about who the Approver is. When five people in a room all think they have a vote, or all think someone else does, you get the death-by-committee dynamic McKinsey flagged. When you walk in having already named the Approver, and you bring a recommendation addressed to that person specifically, the meeting has a shape. Everyone else knows their job is to pressure-test your reasoning, not to relitigate the whole thing from scratch.
I started doing this explicitly during a fractional engagement where a leadership team had been circling the same platform decision for two months. Nothing was wrong with the analysis. The problem was that no one had ever said out loud who owned the call. Once we did, and once the person who owned it started receiving recommendations instead of option menus, the backlog of stuck decisions cleared in about three weeks. The analysis had been ready the whole time. What was missing was someone willing to say “here is what I think we should do” and someone clearly designated to say yes or no.
Why PMs default to the safe version anyway
If recommending is so obviously better, why does the option slide survive? Knight names three reasons and I have watched all three up close.
Some PMs do not believe recommending is their job, especially in low-trust organizations where having a point of view feels like overstepping. Some lack the confidence, whether from thin customer access, analysis paralysis, or plain impostor syndrome. And some have been burned: they made a strong recommendation once, got overruled or blamed, and learned that silence is safer. That third one is real, and I do not want to wave it away. Getting publicly overridden is unpleasant.
But look at what the safe version actually protects. It protects you from being wrong on the record. It does not protect the decision, the roadmap, or the team, all of which are worse off when the person who understands the problem best refuses to say what they think. And over time it costs you the exact thing you were trying to keep. Executives learn which PMs help them decide and which ones make them work harder, and they route trust accordingly. The PM who is occasionally wrong but always has a reasoned position gets pulled into the important calls. The one who only ever presents options gets treated as a research function.
Being overruled with a clear recommendation on the table still leaves you better off than being ignored with a neutral one. In the first case you shaped the conversation and you are on record as the person who saw the problem clearly. In the second you did the work and gave the credit, and the influence, away.
The one change to make this week
You do not need a new framework to start. You need to add one slide, or one sentence, to the next decision you bring forward: Here is what I recommend, and here is why. Keep your alternatives. Keep your analysis. Just stop stopping one step early.
If you want to sharpen the delivery, a one-page decision brief forces the recommendation to the top where it belongs, and knowing when a “yes” from the room is real versus polite keeps you from mistaking silence for agreement. But the mechanics matter less than the mindset shift. The room is not waiting for your options. It is waiting for your judgment. That is the part of the job that does not automate, does not delegate, and does not show up on anyone else’s slide. Bring it.
Sources: McKinsey, “Three keys to faster, better decisions”; Jeff Bezos, 2016 Letter to Amazon Shareholders; Jason Knight, “Why Product Managers Should Always Present a Recommendation,” One Knight in Product; ProductPlan, “DACI Decision-Making Framework”; Itamar Gilad, “The Confidence Meter”.