1/ Perhaps the most important factor in deciding whether to build custom software or use a 3rd party provider is the cost of the "tech debt" on your company's balance sheet.
2/ Most people don't think about it this way, but when you build custom software, you are effectively buying a rapidly depreciating asset that's going to have to constantly be repaired.
3/ At the margin, it's almost always better to use a 3rd party software that's not quite as customized as you like b/c what you lose in personalization is more than compensated for by moving the tech debt onto someone else's balance sheet.
4/ @zackkanter got me started thinking about this after this piece on serverless came out: techcrunch.com/2018/12/15/the…
But I think it's a helpful principle that's more broadly applicable than serverless.
But I think it's a helpful principle that's more broadly applicable than serverless.
5/ We're working on using more no-code solutions like @airtable and @Formstack because they offer a nice tradeoff where you get most of the personalization of custom-built software but also push most of the tech debt onto their balance sheets.
You can also generalize this beyond software, it often makes sense to hire 3rd party service providers because you can effectively transfer "SOP debt" onto their balance sheet.
Applies to Website CMS too.
A less egregious but still problematic version of this is doing too much customization.
E.g. Taking a good out-of-the-box WordPress theme then trying to get too clever.
A less egregious but still problematic version of this is doing too much customization.
E.g. Taking a good out-of-the-box WordPress theme then trying to get too clever.
https://twitter.com/photomatt/status/1377021082092244992
• • •
Missing some Tweet in this thread? You can try to
force a refresh
