⛔ Stop building your product analytics in-house
I've watched this happen at least a dozen times now with teams we talk to: customers start asking for analytics, the team says "how hard can charts be," and 2 quarters later they've shipped a worse version of a BI tool and their actual roadmap is on fire.
The problem is that customer-facing analytics looks like one feature but it's actually 5 products stacked on top of each other:
A query layer against production data
Charting that doesn't look terrible
Permissions and multi-tenant isolation (get this wrong once and you're writing a security disclosure)
Performance that holds up when a customer asks for 2 years of data
And now, because every customer expects it, an AI analyst that actually gets answers right
Each of these is somebody's entire company. Your team is going to build all 5 as a side quest, and then maintain them forever.
I'll admit my bias upfront: we sell BI software, so obviously I'd say this. And there are real exceptions, like when analytics is your core product, or your data model is genuinely weird. If you're Amplitude, build it (obviously).
But for everyone else, the build decision usually comes from engineers underestimating the boring 80% (permissions, edge cases, "can you export this to PDF") rather than any real strategic reason. The fun part, the charts, is maybe 2 weeks of the work (or maybe 2 hours now with Fable).
We felt this so strongly that we turned our whole product into an API for exactly this use case. Launching it on Product Hunt today, take a look!
Replies
@maxmusing - This is going to be a gamechanger for Basedash. I've been contemplating this (using Monaco code to build a full SQL editor within Salestrics), but what you're doing is actually pretty cool. I'd love to chat about this further to see how you could help us implement this within Salestrics.
We're launching Salestrics AI today as well. Check us out.
I like that you acknowledged the bias. It makes the argument more credible. One thing I'd add is that customer-facing analytics also becomes a trust problem; Once users make decisions from dashboards, data freshness and correctness become product features, not engineering details.
I had an exact opposite - lots of money and effort to build an integration with one of the top BI tools only to learn that it's often unstable, hard to customize and needs advanced sync and re-sync handling and message delivery on our side. In addition it turned out that big imports had to be fully manual. And because the company has already spent too much - our CTO wouldn't want to even hear about dropping that tool.