I think this is a great idea for a product, but in my experience, the pain associated with handover is something that is earned over a long period of time. Knowledge starts to centralise in one person and they don’t do enough to disperse it. Maybe they are too busy, maybe they don’t value documentation, maybe it’s an oversight, there are lots of reasons. But that problem starts to build up and you only discover the full extent of it when they leave and it hits you all at once.
A better solution to this problem would be to improve detecting this problem so you can address it in its early stages instead of when it gets bad. One way of doing this is mandatory time off. I don’t recall who said it first, but time off is chaos monkey for people. It’s common in some industries for employees to be forced to take time off to intentionally cause their normal tasks to be performed without them. For instance, banks use it to detect fraud, because the same person doing the same task all the time can make it easy for them to cover crimes up.
If your team is stuck in a rut and the same person is always responsible for doing important tasks every time, this is a risk. If they took some time off, this would help you figure out what knowledge is centralised with them. Don’t put their tasks off and wait until they get back. Figure out how to do it without them. If you struggle, that’s a clear sign you’ll be in for pain when they eventually leave for good and shows you what knowledge they need to disperse.
I once quit a job and gave them 6 weeks notice because I wanted to try and make the handover as painless as possible. My manager said “stop working today. Come in, check your emails, but don’t do any work. Forward and delegate everything that lands in front of you. Let people who reach out to you know you’re leaving and introduce them to whoever is taking over that particular responsibility. We’re going to have to work out how to do things without you. We may as well start today, safe in the knowledge we’ve still got you in the building for the next 6 weeks if we can’t work anything out”.
You can debate the cost effectiveness of the approach, but there was no doubt that by the end of those 6 weeks they were going to be just fine without me.
That's a really interesting approach. Do you think that was too extreme? Like you could've at least done some documentation work in that time perhaps, or that it went well enough that you'd approach it that way again in the future?
There are two parts on this approach. The 1st one is the positive "hey, chillax, transfer knowledge, don't be stressed, etc." The 2nd is security: "this guy cannot be trusted to 'touch' anything, we don't want him to plant a timebomb, we don't want him to mess things up, etc."
It's not 'polite' to bring up the 2nd, but it _is_ there. Imagine me/you/him/her doing something wrong and bringing down Prod for X hours. Then imagine your manager telling the director "oh, and btw he submitted his resignation 2 weeks ago, and in confidence he told me he goes to our ABC competitor". That-Looks-BAD, even if it was the most honest of mistakes and even someone else's fault. Once you 'stop playing', don't touch that jenga tower!!
Sure, but writing documentation doesn't touch the Jenga tower. :)
Happened to me back in 2009 and I couldn't believe it. I got paid for 1 month to sit around, drink coffee, answer questions. Then I moved and got a 30% raise.
This 1 month also helped me refresh all my knowledge, go through my notes, etc. "In teaching I learn, and in learning I teach". Having around 100 people asking me questions from all over the 'region' helped me put things back together, see connecting dots that I missed before, and I left there with a new (better) set of notes about myself and how to do things.
I’m the kind of person who is conscientious enough to make a handover document when I go on PTO, and the same traits that make me do that are the reason why I’d never use this. It lacks a personal touch; in fact it’s almost insulting (“I couldn’t be bothered to prepare a handover document so I asked this AI to do a shoddy job of it”). And the whole point of a handover document is to capture top-of-mind priorities that exist nowhere outside of your own head. An LLM won’t fix that (at least in the near-term before we’ve all got brain implants…)
Same on longer PTO doc, because there is always something weird going on that people should be aware of and no one should be able to say 'No one told me.' From that perspective, it is a useful CYA, but in my experience it genuinely helps everyone plan a little better.
That is the reason I think the Handover project has merit. What I personally hate however is the format ( video ), because I can see corporations adopting this out of sheer laziness. In other words, I don't wish this project be adopted by corporate America based on personal preferences, but I can kinda see them jumping on it.
Forced vacations are super common in the finance / wall st industry for this reason.
Some larger companies do this in disaster drills or other emergency preparations as well. You give the team the task to stand up a database cluster from scratch and pull a dataset into it - and none of the seniors may be contacted. This is pretty effective at rooting out such knowledge silos. And it can build confidence in the other team members once they manage to do it. The main drawback is that it's just non-revenue generating preventative work.
As an IT/Admin/SRE team, we've also found that training isn't just good for the industry, but it also helps to keep such knowledge islands in check. I currently need to make sure that my junior is able to take over most crucial daily tasks, and in the future, all of my important technical tasks - and he's doing wonderful. Though that's now a staffing thing, which costs money.
In the end, it's a question how sustainable a company wants to be long-term I think.
I run a company with almost 100% of the staff located in Europe. Because of this, everyone has 38 days of PTO with some people taking even more unpaid time off. This means everyone is always out of office. We also have a very significant amount of process documentation, and like you alluded to, means significant resiliency. When someone is gone, the work carries on as normal.
It's really only a large risk if you don't compensate competitively.
Sports teams constantly "lose talent" because they refuse to pay for it. Suddenly they start losing games..
All this "knowledge lost" is often just talent and experience.
You can lead a horse to water, you can lead a junior to documentation.
I know the manager brigade will take this personally but talent and experience have always carried any workplace I've been a part of.
Compensating competitively might stop somebody from leaving for another job, but it won’t stop them from keeling over dead from a heart attack or getting hit by a car.
You need to plan in advance for people leaving, not just assume they will be there forever.
Or moving, or just wanting a different job. Money helps, but it doesn't solve everything.
[dead]