Disclosure: I am a founder of https://slab.com that is also addressing this as a scalable SaaS solution.
There's also ops and support. Building something meant for a dedicated team to keep running is way different from something a team will do ops for as one of many things. A complex architecture (even if contained in a single VM) requires a lot more support from more capable support people on the developer's side as well.
You have to think through a lot there. It's not something to take on without careful deliberation but if you're sure you'll eventually need to do both you should strongly consider it. If nothing else make it part of every design decision and pay attention to operational overhead by your own teams because those problems might end up being customer problems.
Building Saas-first or on-premise-first leads your product toward distinct attractors (features, behaviours, compliance, etc.).
I see a lot of startups creating internal software themselves (e.g., CRM's, email marketing systems, etc.) with a similar rationale based on "values and principles", where an off-the-shelf solution would be a much better use of resources.
Ironically, Stripe's core product is one of those off-the-shelf solutions that seems to be a good fit for a lot of startups, yet some still choose to do things the hard way.
Stripe landing pages almost always get tons of support on Hacker News. This isn't because they're just a good payment processor -- it's because their landing pages are unlike anything else in the industry. They could have slapped together a static site in Jekyll with custom styling and called it a day. Instead, they labored over a landing page that feels well designed with immaculate attention to detail.
Did they have to do any of this? Not really. But it shows on the outside -- and it works. Given that the world is filled with design that just copies Apple and Google, supporting creative R&D is a huge plus for them.
Internal tools are the best way to do that -- they let people experiment and learn without fear of consequences on the outside.
I don't want to be too cynical but it can function as a kind of elaborate PR and recruiting tool, as well as generally support the impression of the company being the Next Big Thing. As Square had to open up their finances and work on profitability, they got more focused on core product. I wonder if Stripe will follow.
I also wonder if there's something particular about payment processors.. maybe overcompensating? Is it part of the business model to kind of puff up a company around what should otherwise be a commodity-priced core service?
And most of that should be fairly autonomous. The fat will eventually be trimmed. Just like at Etsy.
Whether or not this is intentional, it's working that way. I've heard more guests on techy podcasts that work for stripe than google, maybe just because they're so enthusiastic about the culture that they always seem to bring it up.
Maybe it depends on the size/stage of the company to some degree. In the startups I've been involved with (< 10 people), developers have little - if any - time to experiment with things that fall outside the core product.
Company hackathons (as mentioned in Stripe's case) can be a good outlet for that kind of creativity, but the kind of quality you mention ("immaculate attention to detail") doesn't come out of hackathons. The company's leadership still has to make those things a priority (which, by definition, deprioritizes other things).
It's not at all surprising to us that both Stripe and Square have spent a lot of time and money on this problem to support their quick growth past 1,000 people.
“Reinventing” implies this has been done right for Stripe (or any such company) elsewhere.
better use of resources
Effectiveness in complex environments is not necessarily efficient.
I've also seen companies drown themselves in custom ERP systems to the extent that it actually impacts their bottom line.
Point is - if these types of projects can produce an ROI and improve the bottom line, then by all means do them. But know what you're good at - and better, know what you're not good at.
Possibly OT, but heck, they do it with AWS bills that coincidentally often track pretty well what it would cost to colocate bare metal and employ sysadmins.
Why can't Stripe do the same thing? It's got this gigantic customer base of fledgling businesses; why not start building internal tools and then, once they're mature and battle-tested internally, offer them publicly to the growing businesses they already work with? It's an ingenious move.
"Wright continued to develop the editor for [Raid On Bungeling Bay] as a personal toy because he enjoyed it so much. He researched urban planning and realised that others might enjoy constructing and building cities themselves. The result was a more advanced simulation that eventually became SimCity."
- the Wikipedia page on RoBB, which also notes that it selling about a mullion units in Japan was a large part of what gave Wright the financial freedom to build Sim City
I don't know if you would find a good off the shelf product that meshed with Stripe's vision. Some of the discrete pieces, document sharing and search, maybe.
My hypothesis is that these companies are overly obsessed with defining a “mold” of the type of person that belongs at the company, and if you slightly deviate from that mold then you’re disproportionately punished for it.
What breaks a company out of commoditization is often culture and so companies will go to great length to demonstrate their differentiator.
(as a former founder and hiring manager at a bigger company who has spent a lot of time doing this in both settings)
Only certain companies though are able to execute and maintain this type of stuff, and even then it’s still questionable whether it’s worth it or not. I would love to see where Home is in 2 and 5 years.
I worked at a company that used its own homegrown timekeeping and HR tool. Over the years it became poorly maintained as people came and gone and the company (and its needs) grew. Eventually it was decided to jump ship to an "off-the-shelf" product and everyone loved it because it actually did what it was supposed to do.
I highly doubt employees are going to feel more loved because the company decided to build their own wiki and address book.
If it's helpful context, Home has been around for two years and People (its precursor) was made in 2014. We now have a team helping shepherd it along -- you're right that an staple is curating so it doesn't become bloated. We intend to continue to share updates as we learn more, so hopefully others can learn from what works (and doesn't!) as well.
Is the thing nice? Sure it's nice. But I think it's a mistake to extrapolate too heavily from that to other areas of a company's operations. Every tech company I've ever been at has had "dirt floors" in some form or another- a really shitty area that was paid little or no attention so the adjacent cathedral could have another layer of gilding added.
Even "awesome" companies have areas you would not want a job working in. If you were in such a situation, would you want your lot to be improved, or would you want to watch cutesy videos of people who work at your company you will likely never meet? Imagine being in such a position, raising the issues you are struggling with every day to whoever will listen, and being ignored in favor of resources getting devoted to a cool UI for editing out-of-office days. That would probably be, you know, a bummer.
Now contrast that to Stripe who's actually expending cycles and real labor to solve what they've deemed a real problem or at least a tangible goal for their org.
Stripe is an awesome company, I can only hope the rest of the industry will take notice of their culture and process. But even if the rest of the industry doesn't take notice Stripe will continue to attract better talent and get higher quality output from their people because of this type of care and consideration.