Getty Images/iStockphoto

Dresner: Analytics trends include prioritization of BI

While generative AI is the BI craze of the moment, increased investments in BI, attempts to expand BI use to business users and embedded analytics are also in vogue.

While generative AI is the dominant trend in analytics, it's not the only one.

An increased emphasis on using analytics to inform decision-making is also an ongoing trend, according to Howard Dresner, a longtime industry analyst and founder and chief research officer at Dresner Advisory Services.

He noted that the economic uncertainty brought on by the COVID-19 pandemic demonstrated the importance of analytics. Worldwide events since then such as the war in Ukraine, fears of a recession, rising inflation and repeated supply chain disruptions have served to further show how analytics can help organizations wade through uncertainty.

As a result, more organizations are using analytics as part of their decision-making process, according to Dresner's research. In addition, those that already rely on analytics are increasing their investments in BI and data management tools to hone and expand their analytics operations.

Meanwhile, another analytics trend is the never-ending effort to expand analytics use within organizations beyond just a small group of data experts, which Dresner terms information democracy.

BI use within organizations has hovered around a quarter of all employees for most of the 21st century. As enterprises attempt to make analytics more widespread, they are deploying self-service tools, embedding data within the workflows of employees and adopting data management strategies such as data mesh that decentralize their data.

In a recent interview, Dresner spoke about the ongoing trends in analytics.

In Part I of the discussion, Dresner delves into trends beyond generative AI, including the heightened importance organizations are placing on BI and some of the different methods they're using to increase analytics use. In Part II, he goes deeply into generative AI.

Editor's note: This Q&A has been edited for clarity and conciseness.

Generative AI is of course the dominant analytics trend right now. But beyond generative AI, what is a major ongoing analytics trend?

These conditions are somewhat unprecedented. Current management teams have never experienced these things all at once, so it's really challenging. Data is their friend.
Howard DresnerFounder and chief research officer, Dresner Advisory Services

Howard Dresner: Organizations are dealing with a bunch of external forces all bearing down on them. There are a lot of headwinds -- economic, geopolitical, supply chain issues, staffing issues and many more. Just like during the COVID-19 pandemic, organizations that have their data act together will fare better than their counterparts that don't. The pandemic was a dress rehearsal for what we're dealing with now.

Organizations that strategically use information will be better off. We call it being hyperdecisive -- being able to take all of the relevant inputs and quickly transform them into decision-making insights that they can execute on. Those organizations will be much more successful over the coming months and years. These conditions are somewhat unprecedented. Current management teams have never experienced these things all at once, so it's really challenging. Data is their friend.

How are organizations responding to the current economic and geopolitical climates?

Dresner: We are seeing organizations, for the most part, increasing their budgets for BI and analytics as well as performance management. At the same time, they're saying these are critical areas for them. Data and analytics followed by things like digital transformation and performance management are the top three [critical areas of investment for organizations].

They're integral to each other -- you can't do one without the other. You have to do all three simultaneously to be successful. At a macro level, that's what we're seeing going on right now.

With prioritization of data and analytics now a trend, what capabilities are organizations investing in?

Dresner: It's sort of like having a balanced diet. You have to have fruits, vegetables and proteins. It's the same thing with BI and analytics. Embedded analytics is part of it. One of the things I started talking about in 1993 was information democracy, which is how to get timely and relevant insights to everyone in an organization so they can do a better job and align with the mission of the enterprise at their level. Organizations that do that successfully are going to be successful.

To do that, you need a combination of things. Self-service BI is part of that. Embedded analytics is certainly important, because if you really want to get to the far reaches of the enterprise, you have to do it in a context that [end users] already know. In some cases, that context is just good old-fashioned pixel-perfect reporting. You have to understand who the user is and meet them where they are. You can't give everyone a dashboard. They're not for everyone. For some, it's embedded analytics. For still others, generative AI might be the right approach.

You have to look at all the different approaches, all the different paradigms, if you want to achieve that notion of information democracy.

How can an organization know who should use dashboards or self-service tools versus who should have BI embedded in their workflows -- how can they figure out which approach is best for which employees?

Howard Dresner, founder and chief research officer, Dresner Advisory ServicesHoward Dresner

Dresner: You have to have competency centers, which are people who are responsible for this stuff. Otherwise, the user is going to self-select. They're going to say they want some cool dashboard. But that might be the wrong thing for them. We don't really want them spending that much time in the technology. We want them to level-set in the morning and go do their job, which may be product-facing or customer-facing. There are people whose primary role [is working with data], whether they're an analyst or a data scientist, and they do spend a disproportionate amount of time with the technology. But that's not everyone.

That's why competency centers -- centers of excellence -- are important. They are people who are charged with understanding how to support the various users that are out there.

Does data governance play into that with guidelines set up to simultaneously empower end users while protecting the organization?

Dresner: Governance has different meanings depending on who you ask. If you ask IT, it means control. That's not a bad thing, because you don't want everyone to have access to everything. Information democracy doesn't mean chaos. You need to make sure that not everyone has access to sensitive information. But the IT approach is very control-centric. If you ask the user, they want reliable information. They want certified copies. They want the reliable set of data that's been fully vetted that they can use, that they can trust and count on. They're not incompatible with each other. You have to have them both.

You end up having a divide where IT says, 'This is the certified approach to using this data, and here's the tool that you're going to use.' The user, meanwhile, wants to use some other thing that they're comfortable with, and they'll do whatever they need to do to get the data. That's where there's a disconnect -- and a competency center can sit in the middle and be a go-between that supports the users and also collaborates with IT to make sure the content is governed in a way that is useful for the end users.

What's your view of data mesh and the trend of decentralized data management and analytics as a way to address that divide between IT and end users?

Dresner: It can work, but that's on the organization.

I like the approach of distributed responsibility. But it can only work if you have a strong set of policies, processes and culture that works well together governing the data. If every [department] is off doing its own thing, it's not going to work. If you have a well-established set of policies, processes and culture, data mesh can totally work. If your idea of mesh is that everyone does their own thing, it will absolutely not work.

Eric Avidon is a senior news writer for TechTarget Editorial and a journalist with more than 25 years of experience. He covers analytics and data management.

Dig Deeper on Business intelligence technology

Data Management
Content Management