Once the project is delivered, make sure you all sit down and do a retrospective on what went wrong, decide what you’ll change next time, and actually make those changes.

If none of this seems feasible in your organisation any time soon, leave. Don’t waste your time with people who aren’t taking your work seriously.

I used to lean that way, but these days I disagree. Sometimes it's worth it to bolt as fast as you can and don't look back to meet a deadline. It's the difference between a $200m raise with a bunch of untested API endpoints and a $10m raise with them.

I'm speaking as the guy that ended up writing most of those missing tests. You don't have a lead for long in some of these spaces and sometimes "strike hard and fast" beats "death in the tarpit of tech debt" in the medium run.

BUT.

YOU NEED TO PAY THAT TECH DEBT OFF AND MOST PEOPLE DON'T THEN THE BUSINESS FAILS ANYWAY.

Sorry for yelling, but that last part is easy to forget.

I'm tired of companies with API endpoints that leak data like a sieve. This is why companies need some skin in the privacy game.

Purposefully neglecting the safety of user data for "speed" like this should have your company taken away from you. Why can't adults stop behaving like toddlers?

> I'm tired of companies with API endpoints that leak data like a sieve. This is why companies need some skin in the privacy game.

There is a world of difference between “untested” and “insecure” API endpoints. What seems to have been cut in the example isn’t a permissions model, but some form of automated integration testing.

I’ve seen horribly insecure APIs with 100% code coverage, and I’ve personally burned myself on untested API endpoints where the privacy implementation was _too_ restrictive for what my customers were trying to do.

It certainly seems like a very successful strategy.

Tell your clients that you apply best (/better) practice.

For this kind of kill point however, you better make sure not just your own endpoints have cleanly passed scrutiny, but your entire modeling and process around them doesn’t produce insecure endpoints. Otherwise it is best not to let your sales team use such a kill point.

Second make sure that list of business requirements is actually a list of business requirements. Often I see clients coming with lists of missing functionalities (e.g. a list with the desired functional design = waterfall). It's nice that clients come up with their suggestions on how things could be implemented. But these should not be set in stone and always come accompanied with the actual business requirements. If the business requirements are unknown the team will be unable to know if certain things might be covered in other ways. Or come up with alternative/reduced solutions to get them covered more quickly.

There are great managers out there, who will take your job seriously, and who can amplify the value you create together, and there are “managers” who know only the whip. If you only know the latter, keep looking because you deserve better.

IMO, this is the only correct answer. Unless you're in a "lose the company" situation, burning out the developers will only make things worse.

Find a new job and move on, take the good devs with you so you don't lose the synergy you built together. Tell them exactly why you're all leaving, perhaps next time they won't have so many damn meetings.

Most importantly - under no circumstances agree to work over 8 hours/day or weekends without them paying you double your usual rate. They must learn that shitty management has its price.

Yes. Don't be me. I pushed myself too hard and got shingles* in my mid 30s, and activated my celiac disease (I've always had the celiac gene, but I find it quite the coincident that symptoms first showed up during this time) all within a 2 month timespan. A very stressful time, and it feels like those 2 months ruined my life, and it all happened so fast. Beware.

* Shingles was weird. Started with a weird feeling in deep deep tissue, then later a surface level rash with itchiness and numbness that comes and goes. There was one evening where the rash had flashes of extreme pain, just two or three times, and I came to realize why people with severe cases describe it as the most painful thing they've experienced. Fortunately, my case was mild and cleared on its own without much pain.

https://www.reddit.com/r/HellsItch/

From many years in film and game development where crunch times were long and brutal, I totally understand (first hand, from experience) how bad this can get. I sympathize with the frustration.

However, be very very careful with these two bits of advice. The first one (taking people with you) is sometimes illegal, and even where it’s not, it will not be taken lightly, it could lead to legal and harsh retaliatory consequences that will not be pleasant.

The second one is also dangerous. Things you say when you leave can follow you out the door to your next job. You probably do not want your reference checks, or just people who know each other, to be spreading rumors that you’re combative and troubled regardless of the fact that poor management is at fault.

The better advice IMO is to find the better job and move on quietly. Trying to teach the company a lesson is very unlikely to work.

As an East Texas proverb states, “Don't try to teach a pig to sing, it will irritate the pig and frustrate you.”

But sure, being quiet about it is an option too - choose based on the situation. And indeed, err on the side of safety.

In most of the world (and where I live) it isn't, on the contrary having anti-poaching policies is illegal and resulted in a huge anti-competitive/wage-supression lawsuit:

https://knowledge.wharton.upenn.edu/article/silicon-valleys-...

https://www.classaction.org/blog/silicon-valley-antipoaching...

What's even more egregious is that a group of people leaving to start something better together is part of how silicon valley got started:

https://www.wikiwand.com/en/Traitorous_eight

Do keep in mind that this case (companies agreeing they won’t hire from each other) is a bit different from an angry employee leaving and soliciting some friends to go with him to a new company. Like I said, it might be perfectly legal, and yet still cause you serious trouble if you do it. Some people have ended up a lot more frustrated after leaving because of things they did on the way out. It can work out well if you’re really in a position to start your own company and fight like the Traitorous Eight, but for most devs who just want a job, making huge waves on exit is pretty dangerous. Don’t take my word for it though.

> but for most devs who just want a job, making huge waves on exit is pretty dangerous

Yeah, it is easier (both in effort and in future prospects) to keep a low profile. Though I was focusing on legality/acceptability of poaching/leaving as a group.

At least in my experience people agree to that, and it's usually when exiting a company, not when joining. As in "you agree not to initiate poaching up until 12 months after leaving the company".

Where I live that can be legal with the following caveat (same for non-compete):

- They have to pay you (a consideration) when signing such agreement

- It must be for a definite period of time. (Unsure about how much time is legal though)

- The only penalty for breach would be to return the consideration.

Where I live most non-competes/non-poaching agreements would be unenforceable since they don't meet the previous criteria. (Unenforceable clauses are normally refereed to as illegal though)

Does anyone even check references anymore in tech? My last two jobs haven't. The most current one didn't even ask for any. I figured that they realized that candidates won't list someone who will provide a bad reference, so it's a waste of time.

There are also informal “reference checks” though, where people who know each other talk about you, and even these are just one of many ways that your reputation can follow you. Leaving behind something very noteworthy like telling your bosses off and taking other employees with you is a good way to have people sharing stories about you.

The rewards have been nearly fully allocated into upper management caste over the past generation or so, but with no carrot they can't use the stick as much as they used to.

But the delivery time and efforts will become the new baseline standard.... to be exceeded in later project of course.

Sometimes, to save a piece of wood, the rot needs cutting out. Similarly, to improve an organisation, bad projects must be allowed to fail.

So I took the existing plan, from before the latest reality happened, and:

* Turned plan into refreshed sorted spreadsheet rows of prioritized tasks and estimated durations and assignments.

* Culled tasks very heavily and sometimes painfully.

* Kept in mind that the system had to work immediately at launch, and had to keep working, with production line uptime and correctness. (For example, there were some non-obvious system robustness/resilience tasks that I added, even as I was removing most other tasks.)

* Looked for opportunities to do things smarter, from the remaining essential tasks.

* Made sure the sick people weren't assigned to anything that absolutely had to happen.

* Kept updating the plan, sometimes multiple times a day, as tasks were completed, problems found, etc.

* Made sure that the estimates always said we'd hit deadline.

All this planning was rapid and intense, not "let's take a day to plan, and schedule some meetings, and have regular check-ins, etc.", and captured in a single view (the spreadsheet).

It was successful, ended up with 100% software uptime and correctness, for the entire contract period, of over a year. Some of that was because the technical cofounder who'd done all the initial work had experience with critical systems, and I followed through in the same spirit. Some of it was luck. Some of is was being vicious and smart about what work was done. Some of it was misery, knowing that the company, and a lot of people's livelihoods and career goals, relied on this coming together in time and working.

As for how to shield people from ulcers in a crunch: if anyone has to get ulcers, it should be leadership, quietly, from figuring out how to make this come together successfully, without the entire team exploding with ulcers.

And the team should see that leadership is all-in on making this be a success, and every individual implicitly buys in on staying committed to the team and project, and the team rises to the smart things that they are asked to do.

(Note: In this case, I used a spreadsheet, but in other cases, the single source of truth view for the entire project would more likely be a good Gantt diagram, or maybe a Kanban board. What's important are that it always reflects the best information and current decisions&activity, everyone is working from it and understands their parts in context rather than as checkoff tasks out of context, and it shows the project coming together.)

>

> * Culled tasks very heavily and sometimes painfully.

This is how any software project should run, continuously.

1. Some things done in everyday practice have big inefficiencies, or give the illusion of productivity and progress, and desperation can force a team to realize that.

2. A lot of tools are driven by enterprise sales, which means they target needs of large companies with very different needs than a startup-like team (in which it's easier to have everyone on the same page, aligned, and barriers removed).

3. A lot of tools are driven by enterprise sales, which can mean that the people who select and approve them aren't the people with experience using them. (And sometimes a team is forced to use poor tools, with adverse impact on effectiveness and morale, and then you get into a budget meeting, and see the SaaS fees the company is paying for those tears (TaaS).)

4. There are definitely projects I can't handle in a spreadsheet (most of them). But, when SHTF, and priorities change, the number of things you're focused on gets smaller and more in everyone's heads, the spreadsheet or text file might be most rapid to work with. (For SHTF deadline-hitting in another, much larger, startup, I would've forced everything into a canonical Gantt chart, just because the interdependencies were too complex, and declared the tons of Jira sprint Issues and manual team reports to be confusing and time-wasting noise, and everyone works&reports first from the Gantt.)

If not, and your company asks you to work your ass off for their screwup, leave and never look back. You might be pleasantly surprised at what developer compensation is like these days.

If you are in a position to influence things, it depends how badly off track you and your team are. You need to be able to push back on unrealistic expectations. If you work yourself and your team to the bone and you still won’t achieve what’s being asked, definitely don’t bother. You’ll still have unrealistic and unreachable targets and all you will have achieved is moving the needle on how much you’re expected to work. If you think you might be able to pull off some heroics and save the project with hard work, you need to ask yourself: “Will I personally see any benefit from doing this? Will my team?”. If the answer is no, don’t do any heroics. Your bosses wouldn’t do it for nothing either. They may have KPIs that depend on this that you don’t have. That sucks for them.

You don’t want to set a precedent that you will pull weekends and evenings if you won’t see any benefit from it (think hard even if it has some upside).

What should happen is that the scope should be reduced to something that can be delivered in the original timeframe if possible, and then some milestones should be moved further out (far enough out that it’s deliverable). You don’t want to buy yourself time and then find it’s not enough.

Did I get anything extra for it? I don't know. I can't point to a specific check that was tied to that cot. I do know that the company appreciated it and my overall trajectory there was good. I was also proud for myself that I could dig deep and really "bring it" for several weeks to reach a goal that meant something to me.

If I had to do it every year, it would grate on me. If I had to do it because other people screwed the pooch, it would annoy me. But I still look back fondly on the overall experience there.

I've never had a better relationship with clients with another model. They get to understand more of what a real development project is. We even invite them to our Cycle Demos for their project.

The best way to not get behind is to not set a deadline. Instead set a scope and budget. Communication is key. More often than not, adding developers does not speed up a project, only adjusting scope does.

Adjust Early. Adjust Accordingly. Adjust Often.

I was in a meeting for a late-running project where there were two SMEs, me as the lone engineer, and seven managers. Project manager, scrum master, product owner, function vertical manager, project coordinator, and two front-line managers... in each daily stand-up.

When I said the number of managers was slowing us down because we had to constantly explain the project, they added a technical manager.

I'm looking elsewhere.

Agile as a methodology solves the issue at a strategic level, but at the end of the day it is always human behavior that determines the way things work. Agile is a tool, and a tool is as good as the people who use it.

I suspect that your project encompassed a lot of people, I wonder if there is not since point of responsibility/accountability, so nobody make the hard choices, and the process was too "democratic".

Nothing can be done about this very project, and this is a symptom of the company as a whole.

This is how companies fail.

I'm not going to say you should never work extra hours but 9 times out of 10 this is a HUGE red flag. The company is responsible for managing timelines and setting priorities for what you should work on. I have occasionally put in a few extra hours (I'm talking <10 in a week) in a crunch period but never for more than 1 week in a row. Anything after 1 week starts to become the norm and that's not what I agreed to when I joined. Salarried does not mean they own you.

Weekends are 100% out of the question except for production emergencies (and if there is an emergency every weekend then no, there's not, don't be tricked into that BS). The business can decide to move the deadline or cut out some of the features but they can't decide to just get more hours from you for free (or rather they can but don't put up with that shit, find a better job).

I have a friend who was in this situation and he hadn't talked about it with anyone until he was nearly at his breaking point. He was telling us (we had met up for drinks) that he was working till 8-9pm if not later almost every day, every week and still feeling like he wasn't getting enough done. He was having regular breakdowns and crying at his desk. Sometimes you are in so deep you don't realize how absurd the requests from your company are and they were guilting him hard about it. This is someone I greatly respect and I would have never expected him to allow himself to be treated this way. We set him straight and told him to set boundaries, only work till 5pm, tell his boss to set the priority but he wouldn't be working overtime. He was in a contracting situation so his actual boss wasn't really involved in what the client was asking him to do but his boss was not a good boss and didn't help him out at all (meaningless platitudes were all he gave, he should have stepped in and set boundaries with the client). He set the boundaries, pulled himself back from the brink of despair, and started looking for a new job. He found one and he is much happier now.

Companies will take (and sometimes guilt you into) every hour they can get from you and rarely tell you to work less. Set boundaries and if they balk then start looking (though it sounds like you should start looking immediately no matter what).

If they don't get that, yeah, leave, will a smile on your face and a song in your heart. The only reward you'll get for staying, if they're not the sort who appreciate the above, is burn-out and more of the same crap in the future.

I don’t see why we don’t skip straight to step 3, but meh.

Once something shipped, if it worked, then no one remembered the pain until the next one.

If everyone is working long hours including execs and it's shared pain, something will eventually change for the better. My problem with it was when I worked long hours but the execs went on long vacations and weren't even available for approvals or questions. Then nothing will change and you should go elsewhere.

And you'll develop a Cassandra-like ability to see it coming when it is coming, and you'll also learn that the absolute worst thing you can possibly do is to try to get ahead of it in any way. Don't point out to them that they're on the road to ruin or they'll blaze down the road faster and then blame you for it.

This is the only relevant point, and you can get your answer by asking if you will receive additional compensation for the additional time that you're being asked to put in. Paying you a salary doesn't give them the right to pretend that _all_ of your time belongs to them. If they want more of your time, they need to pay you more. If they say no, then you say okay and either stay and work exactly 40 hours per week until they fire you for "not being a team player" or whatever (for which I believe you would have cause to seek unemployment benefits, but I am not a lawyer); or leave.

Putting more pressure on developers is the _least_ effective thing that management could possibly do. Stress produces cortisol. Too much cortisol results in overall lower cognitive ability. Why would you stress out the people who need to be able to think clearly?

If you actually succeed in saving the project they’ll reward themselves for their effective motivation.

I’ve had projects that managers refused to staff appropriately and tried to use my pride to work harder and prevent it from failing. I am prideful but there is a limit and they found it.

If it’s happened before, who benefits from this structure? I’m guessing it’s not you (though you may find a way it could be).

If it hasn’t happened before, what’s different this time? Are more or different people or departments involved? What are their backgrounds and motivations?

All this is moot if you don’t have any influence to change either how things are done or how your contribution is perceived; in that case, use this as great fodder for interviews and get a pay bump in the process.

Are you in a management position? If not, there’s probably not much you can do at this point, nor is it your responsibility.

What kind of pressure is being put on you? Are expected to work inhumane hours to meet the deadline or what?

Deadlines are typically not strict unless there is a contract in place. They can be moved. Asking people to put in some extra time once in a while is fine, even if not ideal, but not for extended periods of time. That leads to burnout and the company is valuing delivery over your health and your future ability to work for them.

Was there some kind of thought-up end date? So what, it's running late.

Otherwise a common approach is to cut scope, deliver fewer features than initially planned.

If somebody asks you to work overtime, ask them how you'll be compensated. Factor 1.5 pay for overtime? An extra week of payed vacation? Get that in writing! If the expectation is to do it just to keep your job or "for the company", I'd say start looking for another job.

And no matter what, don't risk burnout. No amount of pay is worth that. If it's too much, either ignore the pressure, or if you cannot, quit immediately.

I kid you not. This is something someone actually said.

Really one off situations are pretty common (once or twice a year). If the company is not making Money - they typically find themselves in this situation for a mad dash (usually results in poor engineering and poor outcomes).

Can you cut corners / steps to make code faster - sure - but if it fundamentally does not change the business / support structure of the company - it will always be the case.

Find another Job.

my current company ( a very large consulting firm ) will pull in people from top to bottom of the project org chart who specialize in getting things back on track. A team is dedicated to relationship management with the client to help them feel more comfortable about the situation, a team is dedicated to delivery management, and very good problem solvers are scattered throughout the delivery team to help with dev and get other devs back on track and unblocked.

preserving a good relationship with your client is all that matters in consulting so we'll gladly take a loss if it means maintaining a good relationship.

I've worked on projects where deadlines were revised each month based on new info, and effort/improvement was focused in gathering better info to inform an (ideally) increasingly accurate deadline confidence.

For fixed deadlines; scope, scope, scope. Cut the scope and take an honest look at why it is late ... start fixing those things with a smaller scope.

I'd never as a team to work overtime, and I work hard to ensure that situation never risks my teams jobs. If it comes to a crunch, the business failed their staff terribly.

Those meetings and processes in the early months of the project? They should have informed both you (developers) and your management about the scope of the project, and how far along you are and how fast you are going.

If you want to change the direction of the project and only if you want you because is a lot of work you should state your demands, negotiate a better compensation and only start after they put their money and compromise on the table, if not, you will get crushed and discarded and frankly is not worth it. My suggestion is that you shouldn't do a retrospective at the end of the project, you should stop it now, if they want to complete the project they should put skin on the game and not just pushing people to work harder.

Very much this. In all likelihood, there won't be time for a retrospective because there will now be many prod bugs to squash, and probably another project or two that is already behind. Or if there is a retrospective, blame will be passed around and nothing will change.

I do work for government tho' so I guess private sector would be less forgiven.

I also want to point out that that other implementation I mentioned was developed, pentested and shut down before production.

Works pretty great.

PS Yeah, you should find a better job. They're out there!

Extending deadlines to remain inside the timeframe.

In the long run renaming late delivery to 'common practice'.

In other companies I've been at the cause was that the codebase was atrocious. Not much to do there if not to keep working as usual. Business kept pulling out resources though.

I have worked at dysfunctional organizations like the one you describe. It's always a communication issue between different parties within the organization. Someone in sales works on a 6-month long quest to sign a new contract that will bring in the money to keep the lights on but it has delivery deadlines inked in that they lie to the team about and tell them the deadline is sooner in the hopes that even if they ship a little late it will still sail in under the contract deadlines. Nobody in the organization has explicitly worked out service objectives and written down the targets and goals so it's always implied that the system should be available 100% of the time resulting in the entire team being paged to handle every frustrated user request. There always several factors that contribute to dysfunction. It's never easy to fix.

If you don't have the political power, emotional energy, or see any possibility where your contributions could steer your organization away from repeating this scenario again, just move on. People are creatures of habit and changing habits takes time, energy, and discipline. But most of all it takes recognition: identifying and admitting that a particular set of behaviours is leading to this problem.

If you can do something about it and feel up to it, it's possible to turn that ship around if others are willing to join in. You have to be willing to step in front of the team, take responsibility, and be a leader. You may get less time for coding yourself and will be focusing on your communication skills and convincing others to join you. It's hard and it can turn an organization around.

At a prior company I worked at this is what I did, I became an engineering manager for a few years to do it, and it transformed the company and the engineers that worked for it. We went from a company that was constantly operating like the one you're describing: always fighting fires, hours of overtime, lost weekends, everyone frustrated. Within a year we hardly ever worked overtime. After another year we were responsibly sharing incidence response, had service objectives, and were shipping continuously. We even took on adapting the organization to take on compliance work in a regulated industry without adding undue stress. Some engineers that came to work with me had never worked on a healthy team before.

However I eventually came to the conclusion that management was too much stress for me and I went back to full-time engineering.

If that doesn't sound like something you want to work on: move on. Unless someone else steps in and does that work you will always be working overtime, being pinged at all hours of the day, wasting away weekends.

Update: clarifying the wizard statement: basically a healthy organization communicates openly and constantly. If we're close to a deadline we've set for ourselves and something isn't going right we talk about it. Often we will move the ship date because we prioritize quality higher than delivering at a particular time. Some times the ship date is important enough that we will scale back our scope and ship the missing parts after launch. But we always ship when we mean to.

If it's typical (i.e., they're comfortable doing business this way) then yes it's time to move on.

If it's more or less a one-off and you believe your efforts will be rewarded eventually - and you're satisfied with the company and culture otherwise - then maybe it's worth staying.

In either case, guard your physical and mental health. In this regard, trust no one but yourself.

If people close to you tell you they get a stressed or exhausted read on you, absolutely trust them. Sometimes we are slow to realize these signs ourselves.

I should have said, "Trust no one at work but yourself."

Thanks for the save.

It is not going to change. We know that it's the worst possible way to run a project. We know long hours are actively damaging to cognitive capacity, introducing extra potential for errors and reducing overall output.

Well, OK, if you were in a position of power, it could change, because you could say "Nope. Not doing that". Short of that?

You can try convincing management that it's the worst possible idea, and they need to either cut scope or move the deadline. This will likely not work, because there's ego invested.

So... do look for that new job, just in case. And get out before you're completely burned out. (And if you can resist at all, make sure you limit your hours to something reasonable. You'll likely start excelling others by week 2)

I think 'projects' in big companies are often DOA because they lose the plot.

Don't get involved in partial problems, but always take flight to where there is a free view over the whole single great problem, even if this view is still not a clear one. - Wittgenstein

The major thing that we found was that you had to look at the whole problem. - Joseph Henry Condon, Bell Labs

... quotes via https://github.com/globalcitizen/taoup

Is the deadline moveable in time or scope? Some deadlines are pretty firm and slipping will result in costing the company a lot of money.

If you stay, you should take notes on how management reacts, what steps they take to improve. Offering the right kind of constructive help to improve it is an opportunity to step into management if you want it. You know a lot more than I do about your situation, but I wouldn’t necessarily assume that meetings were the cause of overtime. Sometimes despite many meetings, lack of enough planning is still the problem. Or maybe the issue is that decisions were not being made in the meetings.

To answer the question, I’ve been in 5 companies that handled this 5 different ways:

- One company (film) that had paid overtime for 2-3 months at the end of every 18 month project. The overtime pay was 1.5x, but some accounting shenanigans made it so overtime pay didn’t kick in until 5 or 10 hours in, so it tended to work out more like 1.3x. Overtime was managed quite well IMO, but wasn’t seen as a problem. Post-project meetings would reflect on how to buffer production from changes in the story or examine accidents and discuss how to keep them from happening. The crunch times over several films slowly went down because management was improving. My base hours were 50/wk, crunch time hours were ~60-70. I did 80 once for a month when a project went off the rails. Management recognized it and this contributed to closing the division and moving everyone.

- Another company (games) had unpaid overtime for 4-6 months at the end of every 18 month project. Free dinner though! Discretionary bonuses tended to go toward people who put in more effort, though it was far from fair. Post-project meetings were slightly more about letting people vent than fixing the planning process, and would tend to reiterate the message that crunch time is inevitable. My base hours were 40-50/wk, crunch hours ramped from 50 to 80, peaking at 90 once or twice. BTW 80 for any real stretch is unlivable.

- When I ran my own startup, my overtime was unpaid and constant, never ending. ;) I probably worked 80-90 hours/wk, but it was way more flexible and much easier to do than when working for someone else’s company.

- When I worked for a web company that used continuous integration and delivery in 2 week sprints there was virtually no overtime, and we would punt features into the next sprint whenever they fell behind or grew in scope. There was a constant mild healthy pressure to deliver, but people clocked out at 5pm more consistently than anywhere else I’ve ever worked.

- At my current job, a hardware company, there is no official overtime or tracking of time at all, including “unlimited” vacation (which is kinda nice but I think tends to make me take less vacation than if I had a quota). Our software delivery is every 3-6 months, hardware projects every ~2 years. I tend to work more than 40 hours a week, and I put in extra hours above that when I want to do a little extra or do a good job on something, or I’m researching something I’m interested in. Compensation is good and my manager doesn’t think of jobs in terms of hours, he and I make sure I’m bringing value over the years.