We were somewhere around Orlando on the edge of Epcot when the excitement began to take hold. The Uber driver was asking how I’d ever got into analytics when we came to a screeching halt far away from the entrance of the Dolphin hotel. I yelled, “We can’t stop here. This is gecko country!” and urged him to park us closer to the entrance.
This was Gartner Data & Analytics Summit 2023: three days of Keynotes, two MDM based sessions, CDO Roundtables, and a whole galaxy of introductions to every data visualization or AI/ML algorithm powered by real or synthetic data. Because once you get locked into attending an analytics conference the tendency is to push it as far as you can…
Day 1 Keynote
Gartner’s inaugural keynote, as always, spoke more than just about latest technology trends or cool new vendors. Emphasis was put on the actual people who will be using data & analytics technology. Gartner highlighted how important it is to overcome skill gaps and language barriers between business users and new data initiatives.
The road blockers to real answers are surprisingly un-technological, Gartner studies show blockers stem from practical problems like talent shortages, aversion to change, and lack of training. Business users are telling Chief Data Officers, “I like your data, but I hate where you have it” and hitting the button for a CSV export.
If you want value from your data and analytic investments these blockers must be addressed the same as any other “request for proposal” check-box.
Day 2: AI and Machine Learning
The stakes felt high for AI and ML vendors all summit long. Gartner has stated, “80% of companies currently use or plan to use “augmented analytics” by the end of the year.”
As one vendor put it, “With AI you only get to make one first impression. If your AI gets it wrong the first time, you are not going to get another shot.” Trust is maybe the most crucial adoption criteria for AI and ML tools. If users can not see the data going into the model they better be able to believe the AI’s answer at face value.
Evaluation cycles are also tricky. When the lifespan of a statistical model is just 3-18 months, many business users have a tough time balancing the amount of time they can invest in evaluating an AI or ML tool before plunging forward with a purchase. AI still seems caught between a rock and a hard place that business users still have to make a leap of faith on fully automated answers.
Days 1-3: Sigma at Gartner
The Sigma team held court in a booth a few stops down from our close partner, Matillion. Over the course of three days 300+ business professionals stopped by to talk about their cloud data warehouses’ analytics strategy. Sigma’s spreadsheet interface made sense to prospects looking to drive value from business users who did not know SQL, but definitely knew spreadsheets.
Sigma also had an opportunity to connect with existing customers. We love to hear first hand impressions of new features like Live Edit and Input Tables. That said, I was also threatened several times that we better not remove our in-app chat support. Support with experts willing to hop on a zoom call at a moment’s notice is tough to come by!
The long awaited Gartner conference was a great opportunity for the analytics industry to reconvene in person and talk shop. AI, ML, and ChatGPT were the buzz of the week, but a lot of folks in attendance bemoaned their data still being on premise. Or even worse, their data being ready in the cloud, but then exported back to desktops!
After visitors came to our booth for a demo and got a Sigma branded 3-in-1 charger, we pointed them to check out our reviews on Gartner Peer Insights to hear what users' experiences have been leveraging their cloud data with Sigma.