Why the Best Software Engineers Are Bought, Not Found

The strongest engineer you hire this year is not reading your job advert.

They are three sprints into something they care about, on a team they like, with a manager who protects their time.

They are not unhappy enough to look, and they will not become unhappy on your schedule.

If your hiring plan depends on good engineers applying, you are fishing in the small part of the market that happens to be available and hoping the best of them turns up.

The applicant pool is not the talent pool

Every job advert reaches the same group: people actively looking right now. That group is real, and there are excellent engineers in it, particularly after a round of layoffs or a bad reorg. But it is a slice, not the market.

The engineers who have the biggest effect on a team tend to be the ones who are least available. They get promoted, they get counter-offered, they get pulled into the next interesting project before they ever consider leaving. Reaching them is not a volume problem you solve with more job boards. It is a persuasion problem.

That is what “bought, not found” means. Not that you outbid everyone. That you go to them, with something specific, and give them a reason to consider a move they were not planning.

Reputation does most of the work before you say anything

By the time a good engineer reads your message, they have often already formed a view of your company. From your engineering blog, or the absence of one. From what your open source contributions look like. From what a former colleague said about your interview process two years ago.

That view decides whether your outreach gets a reply or a swipe. Which makes engineering brand a recruiting asset that compounds quietly. A team that writes honestly about how it builds things, that lets engineers speak at meetups, that has a public track record of interesting technical decisions, starts every conversation ahead.

None of that requires a marketing budget. It requires someone senior deciding it matters and protecting the time for it.

Outreach that gets replies

Most outreach fails for the same reason. It is about the sender. Fast-growing, well-funded, exciting mission, would love to chat.

Messages that work do three things. They show you actually looked at what the person has built, and reference something specific rather than their job title. They name the problem the person would own, in enough technical detail that they can judge whether it is interesting. And they respect that the person is not looking, by making the ask small.

The other quiet advantage is who sends it. A short, direct message from an engineering lead who can talk about the architecture gets replies that a templated recruiter sequence never will. Not because recruiters cannot write, but because the engineer on the other end can tell within a line whether the sender understands the work.

The process is part of the offer

When you approach someone who was not looking, every stage of your process is a test of whether the move is worth the disruption. A five-stage loop with a twelve-hour take-home is a reasonable ask of someone who needs a job. It is an insult to someone who does not.

Compress it. Two or three substantive conversations, one piece of real technical work that respects their time, clear feedback within days, and a decision-maker in the room early. The teams that consistently win passive candidates are rarely the ones with the highest bar. They are the ones who make the decision easy to make quickly.

Speed is not a nice-to-have here. A passive candidate’s interest has a short half-life. The window between “I would consider it” and “I have decided to stay” is often a fortnight, and every unexplained silence in your process closes it a little further.

Pay properly for the hire you say you want

There is an obvious tension in asking someone happy to move for the same money. Nobody leaves a good team laterally on cash alone, and the counter-offer will land the week after you make yours.

Which means being clear-eyed about your levers. If you cannot pay top of market, you need something else that is genuinely true: ownership of a real problem, unusual technical scope, a title that unlocks their next step, equity at a stage where it might actually matter. What you cannot do is offer nothing distinctive and expect a hard sell to close it.

Deciding this before you start outreach saves everyone a lot of time. It also changes who you target. If you are mid-market on comp, you compete on scope, and you go after engineers who are underused where they are rather than the ones being fought over.

Build the pipeline before you need it

The hardest version of this is starting from zero the day a req opens. The teams that hire well have usually been in conversation with people for a year or more. Coffee with someone who said no last time. A relationship maintained with a candidate who was not ready when they first spoke.

Keeping a small, warm list of engineers you would hire tomorrow is unglamorous work with no immediate payoff, which is why almost nobody does it. It is also why the teams that do fill senior roles in six weeks while everyone else takes five months.

Rejected candidates deserve a mention here. Someone who reached the final stage and narrowly missed out is one of the best-qualified people in your network. Most companies never speak to them again.

What to do this quarter

If you take one thing from this, make it the ratio. Look at how your last three engineering hires arrived. If all three came inbound, you are not choosing your engineers, you are accepting whoever showed up.

Shift some of that. Have your engineering leads send five personal messages a week to people they would genuinely want to work with. Cut a stage from your process. Write the thing your team has been meaning to publish about how you build.

None of it is quick, and none of it helps with the role you needed filled last month. It is the difference between a team that gets the engineers it wants and one that gets the engineers that were available.