Great question ​@sarahm!

​@gcfmineo - you might have some great insights on this

​@Senni - Thanks for the tag! You’re not wrong 

​@sarahm - I’ll put the tl;dr up front - it’s a doozy, but the main takeaway should be that the benefits vastly outweigh the negatives, and the product is amazing.

I had no coding experience (outside of hacking together Myspace layouts and messing with some simple Apps Scripts to automate my emails) going into the transition, but the API connectors were wearing us thin. Sometimes we’d max out on call limits, sometimes API calls would stall out during client calls, and sometimes the API would be down for hours. It left us with a lot of wrench icons, bothered clients, and a not-so-realtime dashboard. I looked into data warehousing, tested out GA4’s native connector to BigQuery to see how that kind of thing worked, and I realized our dashboards could have 99.99% uptime, not have to rely on API calls, and if I learned some SQL, we could expand data blending beyond what Looker Studio could offer (a lot of clients wanted to see performance rollups, and we couldn’t satisfy that).
 

At first, and at very least, you get a massive backfill of data that goes back 1.5-2 years, and your data rolls in automatically when you want it. The standard tables Supermetrics offers can more than get the job done for building reports out, it’s just a matter of knowing which tables have what types of data, and building around those limitations. We opted for the Custom Field/Table feature, which was a bit extra on our contract, but allowed us to build exactly what we needed for clients and save a lot of space on data storage. If you wanted to stop here, you still get your money’s worth by having all your data available to you at all times of the day, no limits on API calls, your clients could be halfway around the world checking your reports at 3am, and you can sleep nicely knowing it’s all accurate and ready.

If you’re willing to learn SQL (highly suggested if you go this route), your possibilities are limitless. I started by trying some really simple exercises on W3school and Khan Academy, then once I was comfortable with concepts, had some of the LLM chatbots help with building out some more complex data models. Supermetrics is great about giving enough time when data sources change their output tables, so when that happens, you can code up a new table that seamlessly joins the old ones, and clients won’t see a thing. We’ve exploded our reporting output, going from taking an hour to do reports with the APIs, down to a half hour when we rolled out the data warehouse connectors, and this year we’re launching some AI-enhanced reporting that brings us down to a few minutes of applying a personal touch. Not to mention it helps with joining non-Supermetrics data you might pipe in, like sales data from Google Sheets, csv uploads, and much, much more. Data warehousing with SQL is automation friendly, and the less time you’re spending on the tedious number-checking, the more time you have to bring in new business. 

If you get a bit of sticker shock at the price, yes, the contracts are starkly different and more expensive compared to the API or sheets product. I justified it to leadership as hiring a dedicated coordinator, and with all of the extensibility it’s created and business it’s been able to produce, it’s entirely worth it for our team, plus Supermetrics has a fantastic human support team that can help fix any problem along the way. 

I’m not a paid actor, nor am I endorsed by Supermetrics. They offer a superior product with a lot less headache, less technical debt, lots of support, and are generally great people that I love partnering with. If you have any specific questions, I’m happy to answer them!