Appropriate retention based raises actually require a huge quantity of finesse and genuine ability from line managers and HR.

Even if a company is willing to do so in in good faith, a rarity, a "market price" raise consumes an extraordinary amount of paperwork and HR time and effort.

You take your employee, often parked with a nebulous job title and a role that has morphed entirely from the one they were hired for.

Then you have to define their current role and skill-set exactly, then research the correct band to index that role to in the job market.

For my extremely diverse cross functional team, understanding the market price for the SQL engineer vs the Python engineer in my niche is absurdly complex.

Half the reason they don't do it is because it's just genuinely so much effort to go to. And most of the time the engineer doesn't even appreciate it because they're jumping ship for a change of scenery more than anything.

Every employee you manage to retain in turn reduces HR work however, I imagine even if the sum isn't quite balanced (i.e it's slightly more work to retain than re-hire) then the difference is very easily offset by the massive reduction in disruption to the team in question.

If HR is infact the reason why this doesn't happen then HR is hamstringing the rest of the company.

The difference between companies with high retention vs low retention is massive in my experience, both in terms of code quality and morale. Anecdotally companies with higher retention did better with raises but also generally better with hiring (so they had higher quality employees worth awarding raises to).

>If HR is infact the reason why this doesn't happen then HR is hamstringing the rest of the company.

Yes, exactly.

The thing is you then have to hire a competent HR member to do all this. Believe it or not, the HR industry is actually even more mercenary than the software developer industry.

If you want a good HR employee, it costs money.

All in all it adds up to a "high investment, high reward", approach to hiring and retaining employees. Highly paid HR, highly paid employees, all as pre-requisites to a "good faith" market indexed pay raise.

But for many, many companies, this HR model doesn't make sense, or else management has decided not to implement it.

Sounds like management needs to reduce friction in order to improve efficiency. There is no argument that it is more efficient to lose a successful employee who already knows your business and hire a new one who does not.

A good system would be to always grant the retention raise when presented with a competing offer (without all this work you described), with a short window of time to allow for someone up the chain to veto.

The problem here is the competing offer. By the time someone has gone through the work to obtain an offer elsewhere they have already mentally decided they are making the move. It also poisons the well a bit in terms of recruiting/interviewing as offers come late in that process so if they are being done to farm offers to present for a counter-offer that is likely also happening to your pipeline for other companies, this is bad for everyone.

Retention has to be pre-meditated, you need to identify the market is moving and factor that into your budget. You are -already- doing this in the current system by accounting for market rates to cover attrition. So instead of -having- attrition just pay the retention raises and cut out the middleman and the disruption.

Agreed, let me restate. The least they can do is match / beat a competing offer without reams of paperwork. But you're right, they need to do more than that.