Influence Without Authority Is an Evidence Problem

Every senior engineer eventually runs into the same wall. You can see what the org should do. You cannot make it happen, because the decision belongs to someone else, and the people who would carry it out report to someone else again. Somebody hands you the phrase "influence without authority" and a reading list about relationships, storytelling and executive presence.

The reading list is not wrong, exactly. It is aimed at the wrong problem. In the rooms where these decisions actually get made, the person who wins is almost never the warmest person present. It is the person who arrived with the thing everyone else in the room was missing: evidence that survives being poked at.

The advice that does not survive contact

Relationship advice assumes the blocker is willingness. Sometimes it is. Far more often the person you are trying to move already likes you fine, and is stuck for a reason that has nothing to do with rapport: they cannot yet describe, to the person they report to, why this is the right call and what happens if it is wrong.

That is the actual job. You are not trying to make someone feel good about a decision. You are trying to make it defensible for them. A leader who says yes to your proposal has taken on your risk, in public, usually with less context than you have. Warmth does not reduce that risk by a single point. Evidence does.

This reframe also happens to be kinder to engineers who do not enjoy politics, which is most of them. It says the path runs through work you already know how to do, done in a form other people can pick up and carry.

Three kinds of evidence a decision needs

Almost every stuck proposal is missing one of three things. Naming which one is usually the whole unlock, and it is faster than another round of relationship building.

1. Evidence about the cost of the current state

Most proposals open with the solution. The decider is still on the previous question: is this actually a problem worth spending a quarter on? If the answer lives only in your head, or in a general sense that things are bad, you have given them nothing to weigh the cost of change against.

The move is to make the status quo legible before you make the fix attractive. Pull the incidents. Count the hours, the reopened tickets, the escaped defects, the times a launch slipped for the same underlying reason. Whatever your organization already measures, use that, because a number your leadership already trusts beats a better number they have never seen.

2. Evidence about the option space

A proposal with one option reads as advocacy. A proposal with three reads as analysis, and it tells the decider something important: you went looking for reasons you might be wrong, and you are showing them what you found.

Include the option you rejected that they are most likely to suggest, and say plainly why it loses. Include the cost of doing nothing, honestly, including the case where doing nothing is fine for another two quarters. Nothing builds standing faster than being the person who says the boring option is good enough when it is.

3. Evidence that you can be trusted with the answer

The last kind is about you, and it is the one people skip. Every proposal carries an implicit claim: I am the right person to be believed on this. That claim is settled by your track record of prediction. If you have said before that a thing would break and it broke, and said a migration would take longer than the plan and it did, you are now a source. If your last three confident calls quietly went sideways and nobody ever heard you mention it, you are noise, however good this deck is.

Influence is credit, and credit is issued against a record. You build it by being right in public, wrong in public, and specific enough both times that anyone could tell which one happened.

Authority is a shortcut, and shortcuts run out

Managers get to skip some of this. A director can say we are doing it this way and it largely happens. That is real, and it is also why authority is a poor teacher. Leaders who grow up leaning on it tend to stall the moment they need something from a peer organization, where their title buys nothing at all.

Building platforms taught me this in the least escapable way there is. A platform team has almost no authority over the teams it serves. Every adoption is a voluntary yes from a group that could build it themselves, and the yes has to survive their own planning process, their own leadership, their own quarter. That constraint held whether the domain was content, data, seller experience, payments, connected devices, enterprise applications, e-pharmacy or wealth management. You do not get to mandate. You get to be the option with the least unexplained risk attached to it.

Which is why the engineers who never had authority in the first place are often better at this than the people above them. They have been running on evidence the entire time, because nothing else was ever available.

What this looks like from the other side of the table

Sitting on promotion and hiring panels changes how you read the word influence forever. Candidates describe it constantly. "I influenced the org to adopt X." Almost none of them can answer the follow-up: what did the people who disagreed with you believe, and what specifically changed their minds?

The strong answers are always concrete in the same way. Here is what we were arguing about. Here is what I went and found. Here is the person who pushed back hardest, here is the part of their objection that was right, here is what we changed because of it, and here is what happened after we shipped. That structure is memorable because it is evidence about evidence, and it survives a skeptical reader who was not there.

The weak answers describe meetings attended and alignment achieved. A committee cannot tell whether the person caused the outcome or stood near it. If your own promotion case has this shape, the fix is not better writing. It is that the underlying work needs a trail, which means leaving one on purpose, starting now. That is also why the promo packet is a leadership artifact rather than paperwork, and why the habits that made you a strong senior engineer will stall you before staff if the only record you leave is shipped code.

The three failure modes

First, volume mistaken for rigor. A long document is not evidence. Cutting a proposal down to the one page a busy decider will actually read tests whether you know which facts are load bearing.

Second, evidence delivered as an ambush. Presenting a finished case in a meeting where the owner sees it for the first time asks them to change their mind in front of an audience, which almost nobody does. Walk them through it privately first and let them improve it. The goal is a decision, not credit for the argument.

Third, the missing follow-up. If you never come back and say what happened after the org took your advice, you have spent your standing rather than compounded it. The follow-up is the part that turns one good call into a reputation, and it is the step almost everyone skips.

Influence without authority sounds like a personality you either have or lack. It is closer to a craft with a short list of parts: know what the current state costs, know the options and their prices, and have a public record that makes you worth believing. Anyone willing to do that work can build it, in any org, starting from no title at all.

Haseeb Afsar