Summing Up Three Days at Gartner’s Data and Analytics Conference in Orlando, Florida, USA

How would I sum up several days in Orlando at our 2023 Data and Analytics conference last week – March 19th-22nd, 2023:

• Confusion
• Hype
• Voice of the Business

Fort of all it was a fun time.  You can’t beat getting out and meeting people. It is so real to travel and do stuff.  The world certainly feels a little bit better.

What I Did at the Conference

I had the pleasure and opportunity to present to attendees four times:

• An overview of our modern data and analytics strategy and operating model.
• An overview of one of our D&A Hype cycles.
• A workshop that helps diagnostically map specific data to specific business outcomes.
• Discussed with CDAOs several case studies (here is one) that were used to demonstrate patterns others might follow to better demonstrate the business value of D&A.

I hosted 25 1-1s in between the meetings and presentations.  And I managed to walk around the show floor for a bit.

By about the mid-point of the event, the first overall theme or key take-away started to become clear to me.   It emerged from the 1-1s and conversations with attendees on the show floor.  I validated a few of the causes with demos and discussions with some vendors on the show floor.  There were three big themes for me.  Here is the first.

Confusion

The market is rife with confusion.  I don’t know what it is.  I think that too many words and cute acronyms chasing too few good ideas.  Here are a few areas where confusion exists.

Use cases for a data catalog

Analytics use cases are quite different to governance use cases.  Too often they are conflated.  Several 1-1s asked how to get business folks involved and excited to work with a data catalog in support of a governance program.  That is the wrong question.  For governance, business is already interested but they would only really care about what should be named a glossary.  Even the data dictionary might be used selectively by a steward (in the business) for root cause analysis.  Those are subsets of a much larger catalog.  See Quick Answer: What Are Differences Between a Data Dictionary, Business Glossary and Data Catalog?

Product versus project management

Too many attendees believe that every deliverable in D&A should be a product.  Some even suggest that their D&A platform is now a product.  All data sets, analytics and models should be products and published to an analytics catalog or marketplace.  Second, there is ample confusion between a product and product management.  Products are things.  Product management is a practice.  Not every D&A deliverable is vague and would benefit from an iterative product development approach.  But perhaps now you can see why there is so much confusion.  See Product Management Practices Crucial for Data and Analytics Asset Monetization.

Data mesh versus data fabric

I am not the expert here but in lay terms, I believe both fabric and mesh include a semantic inference engine that consumes active metadata.  Both build semantic maps that span silos of data.  Data mesh additionally requires you to first define information products.  I believe that is a major difference between the two.  But how can these be forecasted with reliability, especially given the point above?  See Infographic: Strategic Comparison of Data Mesh and Data Fabric.

Governance in the analytics pipeline

ETL was where technical folks executed polices approved by business folks.  ETL would execute a transformation that executed a rule, derived from a policy.  This was as good it gets in the world of data warehousing.  When it comes to the analytics pipeline and technology, we should focus on policy execution: how do I execute a policy in this solution?  But too often this is conflated with the business-led policy setting and policy enforcement work. The result is that so many programs are not doing well, and organizations are getting frustrated with the gaps.

Now for the second theme/key take-away.

Beyond Confusion: Hype

If confusion was not enough, the hype and noise from all the technologies and vendors appears to me to be at frenetic levels.  There were so many small vendors, all who seemed to be signaling very similar messages.  I struggled to observe obvious differentiating messages or capabilities.

Looking at the state of the market, I’d guess that cheap capital, searching for yield in the most recent period of near zero interest rates, has flooded what already was a hot market.  It seems 2022 was a record year for VC funding overall.  Apparently 2021 was a record year to that point too: https://www.cnbc.com/2022/01/13/vcs-invested-more-money-than-ever-into-start-ups-last-year.html.  I wonder of much of this money went to data, analytics and AI?  Does the confusion in the market come about because of the cacophony of hype and sales messages from all these vendors?

I visited the show floor a couple of times and was encouraged to see vendors across every segment.  They included data management, analytics and data science, AI and ML, governance and MDM, as well as AI, ML and more.  It was an impressive display.  But is it too much?  Is there demand enough to sustain such a spread?  Maybe.   Some of the confusing points above are clearly tied in with technology hype.  The decision intelligence story is a great example.  While we think it is an important concept, and it does resonate with business roles, it seems that technologists and vendors have different views and understanding of what it really means.

What is a Decision?

During one of my visits to the show floor I was impressed to note the number of vendors sporting “decision intelligence” on their marketing boards.  I excitedly stopped by a couple of booths to ask willing aids to explain and express to me what this meant.  The two I visited demoed a rather mundane explanation of the standard analytics pipeline.  You know the one – it goes something like data discovery, data collection, data clean up, modeling, testing, output, tune etc.

After the demo I asked both, “well, that’s the analytics pipeline of old – where is the decision I am taking?”  After a short, puzzled look, I added, “Is there a way to visualize the decision itself, and how the various elements of the analytics pipeline inform how to improve it?”   Neither offered a visual but both suggested I needed a more detailed demo to understand what was really on offer.  At that point I had had enough, and it was time to move on to the next booth.  See The Future of Data and Analytics: Reengineering Business Decisions, 2025.

Here is the last theme/take-away.

Voice of Business

My third and final trend or take-away from the conference is about the voice of the business.  Much of the conference was grounded in staple advice and examples that are foundational to D&A.  This content tends to focus on communication methods and tactics to show business folks how data drives their outcomes.  This challenge has been with us for many years and won’t likely be gone anytime soon.  It has many names and comes in many forms, and often starts with, “how to…”:

• Demonstrate the business value of data and analytics?
• Speak data?
• Implement data monetization?
• Get and/or keep business involved and interested?
• Implement data literacy?
• What does it mean to become data-driven?

Each of these questions exposes a certain entry vector and set of assumptions.  To ask, “how to get or keep business involved” implies that the person asking is not “of the business”.

How to Shift from Speaking Data to Speaking Business

Several clients shared stories where their actual demonstrations of business impact trumped traditional attempts at creating and selling a business case.  One attendee explained how his multi-year business case for a new D&A platform was rejected.  He then developed a tactic that delivered the platform one piece at a time, one outcome at a time, over several budget cycles.

Another couple of attendees explained how they created data science pilots that showed how very specific business impacts can be achieved, to help tell the story.   In other words, these were small-scale pilots that had a real meaningful impact on a specific business challenge or opportunity.  When this was presented to the business role, it made D&A real.  It was more effective than a math-based business case that is built on a spreadsheet.

Our keynote introduced the value equation as another attempt to weave the threads together into a business relevant story.  The value equation helps tell a story and explore connections between stakeholder and outcomes.  As part of the overall data and analytics strategy, the value equation is a powerful tool to assemble key parts of the story-telling capability.   I myself used my trusty Value Pyramid to demonstrate elements of the value equation in a workshop.  Sets of attendees in teams developed their own value equation.  All are, of course, tied to the wider and overall D&A strategy.

In Conclusion

As usual the conference is hugely uplifting and rewarding.  It is always a massive learning experience.  What cannot be argued with is this.  The audience comprised hungry business and IT folks looking for answers to thorny questions.  Some questions were old, and some were new.  Some were old with new words.  Despite the hype and noise, many attendees were seeking or developing real visions for data, analytics, and AI.  If you attended the conference and are reading this blog, I hope you went home satisfied.  See you in London!