The 2022 year in review

The 2022 year in review

In February I (attempted) to predict the data trends of 2022. Now that 2022 is rapidly coming to a close, it's time to see just how well those predictions fared.

My Predictions Graded

Going for a very American A-F grading scale in which A is perfection, and F means complete, and utter failure.

What I got right:

In-person events make a comeback [A+]

Was it just me or did 2022 felt like the Meetup hay-days of the 2010s? It was such a welcome change after the Covid years of virtual events. Can't wait for more of the same next year!

Evidence that in-person events are back, baby. 

Self-service finds new fervor [A]

Between DBT’s semantic layer, Mode’s latest re-launch, or even just conversations I’m having with customers, there is undoubtedly a renewed interest in self-service. While I have my concerns about the practicalities of many of these solutions, I am happy to see an emphasis on how data is actually used, rather than just built.

Collaboration takes center stage [B]

More tools than ever have collaboration built in, and while data tools are still lagging behind in this field far behind other disciplines in this category, we are slowly catching up.

I started to do some work on thinking through what that collaborative analytics might really mean here.

The Last Mile becomes the next hot topic [B]

I wouldn't say this was as hot as the data mesh in 2021, but it definitely got some significant airtime. In general, people are recognizing all the best-in-class data modeling doesn't mean anything without solving the Last Mile of Analytics. As such, people are starting to think about requirements gathering, quick iterative feedback cycles, and bringing data closer to decision making.

What I kind of got right…

Data teams would be using more and more data tools, all for increasingly niche parts of the data pipeline. [C]

I think a struggling economy stifled the growth of data start-ups we were seeing at the end of 2021, or maybe we just ran out of niche parts of the stack to peruse. But really probably just the former.

Decision scientists are the new Analytics Engineers [C]

So it might not be decision scientists, but there’s been talk of business scientists hitting the data hype cycle. So I'll take partial credit for this one...

Notebooks and data catalogs go enterprise [C]

I’m not sure I’d say they went enterprise, but they definitely went mainstream. The definition of the Modern Data Stack has expanded this year to include not just notebooks and data catalogs, but also orchestration, and data observability tools. But will we see more expansion in 2023?

My bet is you'll see a doubling down in tools (like the canvas 👀) that are focused on surfacing the insights and value from all this investment in the rest of the data stack.

What I got wrong:

Data teams would make mission statements, then abandon them [D]

I’ve not heard of anyone making a mission statement, so maybe they just abandoned the idea before they bothered to build one.

I still don’t think this is necessarily a bad idea, especially after a data team might be coping with a new organizational structure, a much smaller team, and a reduced toolset. So maybe it’s one for 2023?

Python loses its edge [D]

Not even close on this one. Python is surging and not looking back.

Data in the metaverse [F]

To be honest, I’m relieved I got this one wrong.

Stay tuned for some predictions for 2023 in a few weeks!