You had a strong second interview. Everyone liked them. Then the follow-up email went unanswered, and a week later their LinkedIn showed a new company.
From your side this looks like unpredictability. From theirs it was a decision made weeks earlier, for a reason they never told you, because telling you felt awkward and pointless.
This piece is written from the other side of the table. Seven reasons experienced engineers drop out of hiring processes, what each looks like from your seat versus theirs, and what actually fixes it. Rise92 hears these reasons directly while sourcing for global engineering teams.
The context that explains most of it
One fact reframes nearly everything below. Strong technical people are typically engaged in three to five active processes simultaneously, per Gartner survey data.
That changes the question. You are not asking whether they want the job. You are asking whether you are the best of four options, assessed continuously, using evidence you are generating without realising it.
The surrounding numbers:
| Finding | Figure |
| Report being ghosted by an employer in the past year | 53%, up from 38% in 2024 |
| Cite employer silence as the biggest challenge in their search | 59%, three years running |
| Would withdraw over poor communication alone | 47% |
| Abandon a process because it took too long | 57% |
| Believe it is fair to disengage without notice once uninterested | 70% |
That last figure matters more than it looks. The norm has shifted. Silent interview withdrawal is now considered acceptable behavior by most people, which means you will usually not be told why.
1. You were slower than someone else
What you saw: they were engaged, then went quiet after round two.
What happened: another company made them an offer first. Nothing about your process was wrong in isolation. It was just longer.
Time to hire now averages around 41 days overall and runs longer for engineering roles specifically. Meanwhile, top-performing companies deliver an offer within one week of the final interview 64% of the time.
The uncomfortable implication: if you are at the industry average, you are systematically losing to the companies that are not. Brand recognition and compensation don’t reliably offset being third to decide.
What fixes it:
- Measure days between stages, not just total time
- Book the next round before the current one ends
- Hold a standing weekly slot for final rounds rather than finding time
- Decide within one week of the last conversation, always
2. Nobody told them where they stood
What you saw: a gap of ten days while you finished interviewing other people, which felt like normal process management.
What happened: to them, ten days of silence read as rejection. They mentally moved on and invested their energy elsewhere.
Silence is the single most cited complaint in hiring, and 47% say they would withdraw over poor communication alone. Top employers reach a decision and communicate it within three to five days.
The reciprocity problem: when people are ghosted by employers at these rates, they learn that disappearing is normal and do it back. The behavior is taught. Companies that go quiet are training the next group of engineers to go quiet on them.
What fixes it:
- Send an update even when there is no news, with the reason and a date
- Commit to a specific next contact date, then hit it
- Have the hiring manager send a personal note after the final interview
- Reject people promptly and specifically, since it costs one email and buys real reputation
3. Compensation came up too late
What you saw: alignment through three rounds, then a decline at offer stage over money.
What happened: they had a number from week one. You did not share yours until week five, so they kept interviewing elsewhere as insurance, and someone else got there first with a number they could evaluate.
Expectations have shifted here. People now expect salary ranges, benefits, and flexibility details early, and employers who defer compensation to final interviews see measurably higher late-stage drop-off.
Why deferring feels rational and is not: the instinct is that discussing money early weakens your position. In practice, it means spending five weeks of everyone’s time to discover a mismatch that a two-minute conversation would have surfaced in week one.
What fixes it:
- State the band in the first conversation
- Ask their expectation in the first conversation
- If there is a gap, say so immediately rather than hoping enthusiasm closes it
- Treat a mismatch found early as a process success
4. The process asked more than the role justified
What you saw: a thorough loop designed to protect against a bad hire.
What happened: they read a six-hour unpaid assignment or a five-panel sequence as a signal about how the company operates. Bureaucratic, slow, and not respectful of senior time.
An experienced engineer with a demanding current role and other options does not complete a long unpaid exercise. They decline and take the process that respected their time. You never learn they existed, and you conclude the market is thin.
The self-selection nobody intends: long unpaid assignments filter by available free time rather than ability. The people with spare evenings complete them. The people running critical systems at their current employer do not.
What fixes it:
- Cap take-home work at two hours, and pay for it
- Offer a live pairing session as an alternative
- Three substantive rounds maximum at senior level
- For every stage, name what it measures that no other stage measures, and delete the ones that fail
5. What they saw in the interviews put them off
What you saw: a normal loop, no obvious problems.
What happened: an interviewer arrived unprepared and clearly had not read anything. Two interviewers described the role differently. Someone was dismissive about a reasonable question. They were asked to explain something they had already covered twice.
Experienced engineers use your interviews as evidence. They are evaluating how you run things, because how you interview is a fair sample of how you operate.
The specific signals that cost you people:
- An interviewer who has not read the material
- Contradictory descriptions of the role or team
- Questions that reveal nobody knows what the person will actually do
- Any interviewer who seems unhappy
- Repetition across stages, which reads as disorganisation rather than rigor
What fixes it: brief interviewers properly, assign each one a distinct area, and make sure every person can describe the role the same way. A poor hiring process experience is not a soft issue here; it is the only direct evidence they have about you.
6. Internal sign-off stalled after the final round
What you saw: the panel agreed, the offer went for approval, and approval took nine days.
What happened: they accepted elsewhere on day six.
Research in 2026 identifies waiting on internal sign-off as the leading structural cause of losing interested people. It is also the most fixable, because it is entirely within your control and has nothing to do with the market.
What fixes it:
- Get budget and band approved before the role opens, not after the finalist emerges
- Pre-agree who signs and what their turnaround commitment is
- Escalate on day two rather than day seven
- If approval genuinely takes a week, tell the person that upfront so silence does not read as disinterest
The honest test: if your approval chain cannot move in 48 hours for a role you already budgeted, that is an organisational constraint, and it is currently costing you finalists.
7. They researched you and found something
What you saw: enthusiasm, then a quiet withdrawal with no stated reason.
What happened: they read your reviews, asked two people in their network, and heard something that outweighed the opportunity.
Engineers research employers far more thoroughly than most companies assume, and they have better channels than review sites. In smaller markets, one or two conversations settle the question entirely.
What this means practically: how you treat people who do not get the job directly determines your access to people you want later. Interview experiences get shared. A slow, silent, or dismissive process becomes a fact about you that circulates.
What fixes it:
- Be honest in the job description about the difficult parts, since people respect that and disengage less
- Fix the reasons behind the reviews rather than the reviews
- Treat rejections as a reputation investment
- Ask departing employees what they will tell people, and listen
The cost of losing people mid-process
| Cost | Where it lands |
| Wasted interviewer time | Roughly 21 interviewer-hours per technical hire at small companies |
| Extended vacancy | Delivery slips and the existing team absorbs the load |
| Settling | Replacement cost of a compromise hire, commonly estimated near 200% of salary |
| Reputation compounding | Each poor experience reduces your access to the next group |
The reputation line is the one that does not show up in any report and matters most over time.
Where process fixes stop working
Everything above assumes the engineers you want are in your pipeline and you are losing them. Fix those seven things, and you will convert substantially more of them.
There is a separate problem that looks identical from the dashboard. If the senior engineers you need are employed, performing well, and never entering your process at all, no amount of speed or communication changes the outcome. You cannot lose people you never reached.
Telling these apart is straightforward. Look at where your hires come from by source. If your pipeline is almost entirely inbound applications, your constraint is access rather than experience, and the answer is off-market sourcing through closed networks rather than a faster funnel.
The practical sequence: fix the process first, because it is cheaper and it makes every later sourcing investment convert better. Then address access if the numbers still fall short.
What this costs to get right
Almost all of the fixes above are free. Faster scheduling, earlier compensation conversations, shorter loops, and committed response dates cost nothing except decisions.
The two that cost money are paying for take-home work and paying at the top of the band for scarce roles. Both are small relative to a vacancy running an extra two months.
Where cost enters properly is the decision between building sourcing capability internally or partnering for the access-constrained roles. That comparison is worth running on total landed cost including vacancy and internal time, not headline fee, and the breakdown sits in transparent pricing.
Frequently asked questions
Because disengaging silently has become normalised, with roughly 70% now considering it fair once they are no longer interested. It is also reciprocal, since 53% report being ghosted by employers themselves. Expect not to be told, and build your diagnostics around behavior rather than stated reasons.
Speed, in most datasets. Strong technical people are typically in three to five processes at once and generally accept the first credible offer, while 57% report abandoning processes that took too long. Communication gaps run a close second, with 47% saying they would withdraw over poor communication alone.
Yes. Deferring compensation to later rounds is strongly associated with late-stage drop-off, because people continue interviewing elsewhere while they wait for a number they can evaluate. Surfacing a mismatch in week one is a process success, not a lost opportunity.
Fast enough to decide within one week of the final interview, which top-performing companies manage 64% of the time. The overall average sits near 41 days and runs longer for engineering, so matching the average means routinely losing to companies that beat it. Measure gaps between stages rather than the total, since that is where the delay actually lives.
What to take from this
The choice is narrower than the marketing suggests. Establish whether the person is an employee or a genuine contractor, establish whether you own a local entity, and five of the six options usually eliminate themselves.
Then model total cost per country rather than comparing provider fees, because statutory variation dwarfs fee variation.
And keep the sourcing decision separate from the employment decision, because conflating them is how companies end up paying an ongoing percentage for something a one-time arrangement would have covered.
To work through which structure fits your situation, get in touch.
The short version
Experienced engineers rarely leave a process because they were assessed too hard. They leave because something signalled that the organisation is slow, disorganised, or careless with their time, and they had three other options that did not.
None of the seven reasons above requires new tooling or budget. They require deciding that a week of delay is unacceptable, that silence is a decision, and that a senior person’s evening is worth paying for.
Fix those, and your conversion improves immediately. If the numbers still fall short afterwards, the problem is that the people you want were never in the room. To work out which one you are facing, get in touch.
