At my current job (gov. agency) we use this internal tool - think marine traffic combined with flightradar24 - and it is great. Truly awesome.
It's been developed for almost 10 years now, by a relatively large consulting firm.
But I believe a large part of the success comes down to the following:
- The core dev. team is still mostly the same. When you think about it, 8-10 years is pretty much an eternity in the world of software development, so it's pretty incredible that they've managed to retain so many members.
- Development cycles are very fast, and communication between end-user (us) and the product manager is very good. There's very little red-tape or committees between us.
With that said - even though the product is quite extensive, the number of users is quite small, only around 500 active. And they are all gov. workers, though in different departments.
On the other hand, I've also seen some dogshit products in the wild. Usually in the scale has been much, much larger - and the budgets, too.
Observations from those have been pretty much the inverse of the one I mentioned above. Huge, huge teams - always new people. Insane turnover. Getting new features takes years, and everything goes through multiple levels of bureaucracy. By the time something has been implemented, interest has changed.
I'd say recreation.gov by Booz Allen Hamilton is pretty good. Some of the policies are disputed (e.g., racing for camp sites), but the technology works very well.
The large scale consulting firms collectively pull in hundreds of billions of dollars of revenue per year and they're mostly _okay_, you just don't hear about them because nobody writes a front page HN article about "that one time I paid Deloitte $300k to implement an internal dashboard and it turned out mediocre".
I've been part of a handful of these projects. You have to understand that these types of projects are, from the start, never intended to be smashing successes. Nobody says "everything is going great and we have a ton of smart, capable people working here, we should also hire some consultants". No, once your company or project is at the phase where you're hiring consultants, 9 times out of 10 its because shit has hit the fan and you need someone to pull you out of the mud, and quick. Consulting engagements are almost always band-aid fixes just trying to achieve a level of mediocrity.
And as much as people shit on the "24 year olds who don't know anything", the sad truth is that 24 year olds (aka people with 2-3 years experience in consulting that are now on their 8th project doing this exact same type of implementation) may not have much experience, but they still are probably more productive than the vast majority of knuckle dragging drones that work in corporate america. When I was one of those 24 year old consultants, most of my projects were roles where I was supporting "IT admins" who literally had trouble remembering how to open Excel. The people that these consultants are meant to supplement aren't your "average HN commenter", they're not even "average redditor". They're "average facebook poster".
Even the so called success stories seem ... mediocre at best.
Has anyone been part of such a project at a firm like that?