Leveraging a Modern Data Stack: Analytics Tooling for BI Modernization

Leveraging a Modern Data Stack: Analytics Tooling for BI Modernization

Almost twenty years ago, a colleague and I used to regularly refer to Excel as the “super tool” because it seemed like more often than not it was at the heart of business processes for companies we worked with. At the time, there were a lot of situations where Excel did as good a job as any other option.

Excel is an amazing product (still), and it shockingly is still at the heart of more than a few significantly sized companies’ approach to collecting and distributing data and insights. Fortunately though, analytics tooling has evolved to the point where if you’re still relying on Excel, you are well behind the times. BI modernization has enabled better accessibility to more powerful sets of data, with visualization capabilities that bring that information to life.

Proper analytics tooling helps make BI modernization possible

When people hear the phrase “BI modernization,” many first think about data — data in the cloud (e.g., Snowflake’s Data Cloud), data lake houses, clean rooms, etc.) But a modernization of your data stack is relatively useless if you don’t have a way for end users to effectively use and consume that data. Without proper analytics tooling, BI modernization is like a beautiful house (a lake house, perhaps?) with boarded-up windows. A modern analytics platform (including multiple tools — more on that to come) allows all of that data to be consumable and usable and is an essential component of BI modernization.

As part of a move to BI modernization, it’s important to take a step back and understand the features that are available while also determining which are most critical to your business. That said…

Key components to consider as you evolve your analytics tooling

Embedded vs. standalone

For years — decades, really — the only option users had was to “go over here to your reporting system and run this report or view this dashboard.” Modern analytical tools allow targeted, dynamic insights in the context of where people are working. Whether it’s embedding a CRM Analytics dashboard in Salesforce or a Tableau dashboard in a customer portal, your analytics stack should allow you to place insights where users work rather than forcing them to go to the tool.

Actionability

Actionability goes hand-in-hand with the ability to co-mingle insights within a user’s workflow. Yet the ability to take direct action from within a dashboard or insight is often overlooked when evaluating analytics tooling. Your modern analytics stack should streamline actions as much as possible, and the easier it makes it for end users to do the things the insights are suggesting, the more powerful that tool becomes to your business. A great example of this is the pre-built action framework that comes embedded with CRM Analytics.

Distribution of insights

As a user, I shouldn’t have to go and find insights, even if they can be embedded in the context of where I’m working. That’s fine, but what if I don’t happen to be on that screen or even at my desk? As you rethink your BI modernization approach, the ability to intelligently distribute insights is a key component you should consider. Whether that is setting alerts based on certain data thresholds being reached or scheduling distribution of insights, that outward communication of information should be multi-dimensional as well — meaning I should be able to deliver it via email or via collaboration tools like Slack. Ideally, the distributed insight should be interactive; at worst, the ability to interact with the data should be no more than a click away.

Looking backward vs. looking ahead

Perhaps this seems obvious, but if the analytics you’re creating are purely backward-looking and static, there will be a time in the near future where that is simply unacceptable. For leading companies, it already is. Every insight should have a “So what?” — a trend or comparison to help a consumer of that insight quickly make sense of it. But more importantly, it shouldn’t put the full onus of “What does this mean?” on the end user.

That gap can be filled a number of ways, through AI/machine learning predictions (or, better yet, prescriptive insights on what a user should do with the information) or intelligent decomposition of insights into natural language through tools like Tableau’s Data Stories. Regardless, your analytics tools should do more than present a chart and leave a user to draw conclusions.

A modern visualization experience

Most — but not all — platforms support a more modern visualization experience these days. What do I mean by that? They support responsive design across desktop, tablet, and phone layouts; they’re accessible from the web; and they can be interacted with via APIs or web-based integrations. Tableau is the clear market leader in this space, as anyone who has browsed the submissions of Iron Viz winners can attest.

Data recency

While this drifts into the data side of BI modernization, there is a component to consider when selecting analytics tooling that needs to focus on built-in capabilities for accessing data as closely to real time as possible. Things like zero-copy data sharing/connectors, direct queries, and incremental datasets can all be used to provide more powerful insights because the data is as current as possible.

Self-exploration of data

One of the outcomes and major advantages of BI modernization is the democratization of data. If you take the time to shape this and do it well, you can see exponential benefit by allowing end users to start to discover their own insights. Some tools allow for this better than others, with Tableau and Power BI headlining those that do it well.

The most common mistake companies make when it comes to analytics tooling

You’ll notice in the above feature set, I mention several tools. One common mistake we see is companies that try to find one tool to check all the boxes. While some are better than others, there is no single product that outperforms all others across all of the criteria. So what do you do? Do you sacrifice critical features in favor of a lower total cost of ownership for analytics tools? Or do you invest in multiple products to get full coverage of critical features, taking on higher product license costs but also driving a greater return on your BI modernization data investment? 

The answer to that will be organization-specific, but in general the most successful organizations we have seen have typically leveraged more than one tool. To help offset increased product costs, organizations looked for opportunities to very specifically target licensing (not everyone needs more than one tool, typically) and put targets in place to help measure the ROI and incentivize the appropriate usage of those tools to drive meaningful and measurable changes to their business.

How to choose the approach that’s right for your business

In order to make that decision, our advice is to take a use-case driven approach to choosing the right tool, factoring in the features above. Understand a prioritized set of 15-20 use cases across your organization; then examine (and ideally reimagine) the processes that encompass those use cases, and choose the most appropriate tool for each scenario. You may find that one or two tools are the best choice across the vast majority (including all of the top scenarios), and that will help inform and justify your analytics tooling selection as part of your BI modernization initiative.

But what happens when you modernize your data stack and analytics tooling, but no one shows up to use it? Check out our blog post on the importance of business change enablement to ensure you get the buy-in and ROI that you’re looking for on your journey to BI modernization.

Contact us about your analytics tooling needs and how we can help you reach your BI modernization goals.

You may also like...