Published in


20 Predictions for Data and Tech in ’22

Emerging trends to watch for in 2022 in data analytics, BI, AI, and governance.

Photo by Michael Dziedzic on Unsplash

As unprecedented and unpredictable as 2021 has been, the team at SqlDBM is getting prepared for the year to come by trying to anticipate what’s looming on the horizon.

Through conversations with customers and industry leaders, we have compiled a list of twenty predictions that we’ll be tracking throughout 2022 in data analytics, business intelligence, AI, and governance.

Analytics: strategies, tactics, and trends

Whether good (i.e., revolutionary) or bad (i.e., what do we do now), “disruption ”will continue to be the keyword this year. This year will bring more opportunities to use data and new ways to do so.

  • Tactical decisions will dominate as COVID wallops global strategy: As COVID evolves, so should organizational plans and targets. Yearly targets, quarterly reports, even monthly tracking may not be agile enough to react to market conditions. The question is, will your data be able to keep up?
  • Geographic data will go from meta to must (have): To the point above, the geo/regional data component will become as important as the data itself. Snowflake plays nicely with geospatial data, so take full advantage of it in your ELT pipelines and reporting tools.
  • The tech world will shine a spotlight on Data Mesh: as a concept, Data Mesh is well past the idea phase and will be actively trialed by forward-looking enterprises in the year to come. Whether it will be adopted as an industry standard remains to be seen, but it’s already clear that every organization can stand to benefit from some of the ideas that Data Mesh aims to solidify (unless you like your data siloed).
  • Streaming will take market share from incremental data loading: The technology to facilitate near-real-time data has caught up to business needs, and the demand has never been more pressing. Thanks to Snowpipe, Spark Streaming, AWS Kafka, Kinesis, or other solution, the viability of scheduled nightly loads will continue to diminish as streaming technology evolves.
  • Decentralized, distributed data: That decentralized apps and Web3 will disrupt the current finance and tech monoliths is yesteryear’s news. This year, data and the ability to capture it in new ways from previously untapped sources will determine who will be left standing when the dust settles.
  • Data silos will continue to crumble: The Data Mesh, user-friendly tech, and Snowflake’s efforts in data sharing and commoditization are just some factors aiming to democratize data usage. As it stands, besides job security and departmental power games, there remain few excuses to justify the dreaded data silo.
  • Snowflake will continue to exceed estimates: They’ve done it so consistently since their IPO that if I had to bet the farm on just one prediction, this would be it. Edward K, Head of Product Management at SqlDBM, predicts it will hit nine thousand customers by the end of the year.

Business Intelligence — turning information into insight.

No matter what happens in the world at large, the BI sector will continue to grow. Here is what we expect to see this year.

  • Services < software < hardware: Software will continue to eat hardware, and services will continue to eat software. As Snowflake continues to demonstrate, the trend outlined by Forbes in 2019 is only getting started.
  • Beyond ANSI, databases will evolve to meet business needs: just as Snowflake has gone beyond Data Warehousing and expanded capabilities in spaces like Data Lake, Data Engineering, Data Science, Data Sharing, and Data Applications. Modern BI demands it.
  • Forget data lake, forget data lakehouse. Remember the Data Cloud: data lakes were a necessary evil before technology caught up and made working with semi-structured data simple. Thanks to Snowflake’s Data Cloud, by the time we were ready for the lakehouses, we were also prepared to move beyond them.
  • SqlDBM will make data modeling “sexy”: get used to seeing those words in the same sentence. Sure, it’s not hard to improve on legacy modeling tools’ dated look and feel. According to Ajay Singh, SqlDBM’s Product Manager, SqlDBM will move way beyond look and feel and incorporate more functionality (e.g., Oracle support, integration with Confluence, JIRA, Slack, Teams, and ELT tools) than ever. SqlDBM will also add time-saving features like API access, view lineage, and more.
Make data modeling sexy
  • Intuitive SaaS tools will bridge the gap between tech and business teams: Anna Abramova, Tech and Sales Leader at SqlDBM, sees ease of use — see Snowflake’s near-zero-maintenance pledge for a great example — as a significant contributor to bringing tech and business teams closer together. Those that go beyond intuitive, and all the way to “sexy” (see the previous point), will certainly stand out.
  • The world’s data will be roughly 80 zettabytes and rise to 175 by ’25: That’s according to a joint study by IDC and Seagate. If you’re like me, to even begin to fathom such quantities, you’d have to google “zettabyte.” Because orders of magnitude are defined by their predecessors, you’d also need to google “exabyte” before arriving at the familiar “terabyte.” Or you can try an analogy: if each Terabyte in a Zettabyte were a kilometer, it would be equivalent to 1,300 round trips to the moon (768,800 kilometers) — times 80.

AI, the digitalization of human potential

The integration of ML with the organizational tech stack has consistently lagged estimates. Will this be the year that sees broad AI/ML adoption?

  • Analytics will move from explaining the past to predicting the future: COVID has taught us that companies need to learn to pivot in the face of changing reality if they want to survive. Waiting for quarterly or even monthly results to detect changes in customer behavior will no longer suffice. AI and ML integration into a company’s tech stack and the ability to react to change as it happens will be the key differentiator for the businesses that learn to adapt.
  • XOps will be the key to getting the most out of AI: BizOps, MarketingOps, DevOps, AIOps, CyleOps, MLOps, DataOps— so many “Ops” that you may not have noticed the Marvel superhero thrown into the mix. “XOps” is the industry’s latest effort to rebrand and focus on all the “Ops’s” primary goal: seamless, cyclical integration. AI and ML may yield stunning insights, but what will separate good companies from great will be the ability to procedurally re-inject those insights into their tech and BI pipelines and automate decision-making. Register, react, recalibrate, repeat.
  • AI will go from big data to small & wide: before AI came along, BI demanded large data sets and correspondingly large processing power. With AI’s ability to connect data points and pattern match, wider is better for driving insights, as more factors can be correlated.

Data Governance

No matter what happens in tech — new tools emerge, new methodologies are implemented — data governance is there to ensure order, compliance, and usability. And my, does data governance have their work cut out for them this year.

  • Data will get weaponized: according to this Vice article, it’s happening already. Even with anonymized facts, it only takes a few choice data morsels to identify a unique individual with a high degree of confidence. The onus on governance teams to protect customer data is higher than ever. As Grindr found out last year, the fine ($11.7 million) may be a pittance compared to the subsequent loss of trust.
  • New technologies included in the BI stack will be asked to “bend the knee” to Data Governance: as eager as any organization is to try out a new toy, a dose of prudence and protocol is never a bad idea. Putting governance and ethical standards in place before allowing technologies like ML to drive questionable decisions will need to happen quickly (see previous points on tactical decision-making). Some food delivery companies already face serious tickets for speeding in the AI lane.
  • Data Governance teams will continue to struggle to govern effectively: because keeping up with technology is hard. The organizations that get this right will only do so by fostering a culture of collaboration and instilling the idea that governance is a “team sport.” Unfortunately, if the past is any indication, most organizations will attempt to respond by throwing more technology at the technology governance issue.
Data Governance vendors be like
  • There will be fresh opportunities to get governance “right”: External data sources in Snowflake’s data marketplace, the emergence of the Data Mesh, AI, the tearing down of departmental data silos are also opportunities to start fresh and get momentum going on effective data governance. In the words of one innovator: “will you capture it, or just let it slip?”




All about SqlDBM — Cloud based Data Modeling Tool for Snowflake ❄️, AWS Redshift, MS SQL Server, PostGreSQL & MySQL

Recommended from Medium

A Data Science Challenge Towards Better Starbucks Offers

How to Fix Your Data Quality Problem

Automated Machine Learning  — The Next Big Thing

Visual Communication for Data Scientists

Understand — Logistic Regression within 5min 🐱‍🏍

How faulty data preparation will lower your model’s credibility

Tableau meets Einstein analytics! Does it help you?

Data Viz — Chapter 2: Distribution of data — Histogram & Boxplot

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Serge Gershkovich

Serge Gershkovich

Food for thought, meals essential. Shrine your mind, build your temple

More from Medium

Google Data Catalog & Sisense metadata integration

How to Restore Deleted BigQuery Tables

Delete key on a keyboard.

Data Lakehouse vs. Data Lake

Data Build Tool (DBT): SQL on Steroids

Different model layers