ABSTRACT: We enter 2023 in a haze of uncertainty. Here are five principles about what this means for data and analytics.

Economic fluctuations tend to surprise many “experts:” witness the tech implosion 20 years ago, the financial crisis of 2007-2009, and today’s bout of post-COVID inflation. In each case, most economists decided we had a real problem only months after it started.

Our inability to anticipate big changes—and the recessions they often bring—makes 2023 a challenge for enterprises. Executives and the data leaders that support them don’t have reliable advice to follow. Despite this, they must reduce risk and preserve cash while still investing in the right initiatives to maintain a competitive edge. 

So we enter 2023 in a haze of doubt and uncertainty. Here are five principles about what this means to the world of data & analytics.

1. Analytics projects must compete for fewer dollars. “Top of mind is efficiency. Almost universally, budgets will be flat or down due to compressed margins thanks to inflation.” These paraphrased comments came from Jesus Mantas, Global Managing Partner of Consulting at IBM, as he addressed a roomful of chief data officers at a CDO Summit in Boston last month. This reinforces my findings from an unscientific LinkedIn poll in October, in which more than a third of the 30 respondents (sanitized based on role) said they reduced spending plans. It’s fair to say the outlook has not brightened since that time.

Source: My LinkedIn post and survey, October 2022. 

2. Use cases must shift. In uncertain times, enterprises rotate to projects with lower risk and faster payback periods. This holds true for analytics projects as well. For example, fewer finance teams regularly used predictive analytics, AI, or ML in September 2022 compared with early 2022 or one year earlier. Furthermore, most of them expected to invest less in these advanced tools in 2023 than 2022. Instead, most respondents expected to invest more in cloud-based planning and reporting solutions. The implication: traditional business functions are rotating to lower-risk analytics projects this year with faster payback periods. These findings come from an October 2022 survey of 657 financial leaders in North America by Hanover Research. (Note that the survey excluded analytics teams themselves, who might have different plans.)

Source: Financial Decision-Makers Outlook October 2022, Hanover Research.

3. Enterprises must try to control cloud costs. IT departments, and the business teams they serve, can no longer afford cost surprises or overruns when it comes to cloud-based analytics. Many of them will adopt the emerging discipline of FinOps to ensure their analysts, engineers, finance managers, and business owners collaborate to govern the cost of cloud-related projects. FinOps instills best practices, automates processes, and makes stakeholders accountable for the cost of cloud-related activities. And none too soon. Nearly half of 28 respondents (again sanitized based on role) to another unscientific LinkedIn poll, this one in November, said cloud costs limit their analytics projects.

Source: My LinkedIn post and survey, November 2022.

In 2023, look for FinOps to become a required element of cloud-analytics projects, assisted by a rising number of tools and platforms. For example, Unravel’s observability product helps business and data teams forecast, allocate, and optimize the cost of cloud resources such as compute. Yellowbrick’s data warehouse, meanwhile, offers users flat subscription rates and advises them about how to reduce consumption fees from their underlying cloud provider.

4. Data & analytics teams must take a more holistic view of project ROI. Many analytics projects involve multiple use cases, stakeholders, tools, platforms, and datasets. During a downturn, data & analytics teams must reduce the complexity, govern the costs, and ensure the return on investment for these projects. Consider the case of a retail enterprise with more than 50,000 employees, which struggled to assess the progress and impact of ongoing projects. Their data team created a central library of projects, assets and best practices that helped business, data, analytics, and IT stakeholders collaborate in a more efficient way. The data team estimates that its productivity increased 40%. Using a platform from the vendor YOOI, they are gaining visibility into the inputs, outputs, and business results of each project. Enterprises must take similar steps to rationalize how they consume and extract value from data in 2023.

5. Data governance needs more vigilance than ever. Investors, regulators, and customers all become more demanding during downturns. This increases the potential business damage of data breaches, inaccurate records, or violations of privacy. As a result, enterprises must redouble their investments in data governance programs in 2023. Data governance teams must improve transparency and reduce the risk of inappropriate data usage. They must consolidate and reconcile conflicting data sets to standardize on reliable sources of truth. Most of all, they must ensure their communications with customers are compliant, accurate, and respectful. The rise of artificial intelligence, generative AI in particular, makes these objectives all the more challenging because it can automate bad decisions and bad content.

Former Intel CEO Andy Grove is said to have observed that “bad companies are destroyed by crisis. Good companies survive them. Great companies are improved by them.” Whether this holds true or not, 2023 does offer a fresh opportunity to improve how we use analytics.