Proving the Value of Analytics: A Starting Point for Exploration
About 18 months ago, I posed a question on LinkedIn that generated significant interest:
Hard question: If you are an analyst, or you run a team of analysts, how do you gauge whether the insights you’ve delivered have truly influenced business outcomes?
I was preparing for a talk and wanted to see if others cared about this idea. It turns out they did.
I’m revisiting this topic as part of my writing experiment, giving myself 20 minutes a day to write. I'll break this topic into bite-sized chunks, covering it throughout the month. I may intersperse it with other topics—I’m keeping an open mind about where this journey takes me.
Why is it important to prove the value of analytics? In today’s tight job market, with increased scrutiny on throughput, data teams can’t rely on vague promises of adding value. We need to prove it. Businesses fund data teams with the expectation of delivering value, and the onus is on us to show how we do it.
What business outcomes are we aiming to influence? As it has been since the dawn of time (or dawn of modern business, at least) the outcomes boil down to making money, saving money, and keeping customers happy. Every data team effort should impact at least one of these outcomes, though most of the time the proof involves a series of indirect steps rather than a direct path.?
领英推荐
For example, improving customer segmentation might not immediately show a revenue increase. Still, it leads to more effective marketing campaigns, which eventually boost sales and customer satisfaction.
Why does this matter? Proving the value of analytics matters because it highlights the tangible impact of data work. For data team leaders, it means showing that their team's contributions drive business outcomes, which can support requests for resources and new tools. For individual analysts, it ensures that their work is recognized and valued within the organization.
Looking forward: In future installments, I will explore various aspects of proving the value of analytics, including measurement approaches like counting active users, conducting satisfaction surveys, and confirming the actionability of analysis work. I'll also provide tips for leaders and individual analysts on how to make value-proving a systematic practice.
Credits: This series will be loosely based on the talk I gave in 2023 at the Marketing Analytics Summit (thank you, Jim Sterne , for the invitation and inspiration). If you want to watch the full talk, it’s available on YouTube (thank you, Kelly Wortham , for the opportunity to record it with and for Test & Learn Community (TLC) ). And finally, thank you, Lior Solomon , for suggesting that I cover this topic during my July writing experiment.
Stay tuned for more insights and practical advice on proving the value of your analytics work.
Founder, Test & Learn Community (TLC) & Organizer, Experimentation island (#Ei2025) | Founder, Forward Digital
4 个月(side note - really love that you're doing this series and the writing is great and important! Keep writing!) It's funny how so many initiatives that have the *ability* to measure ROI are expected to do so - but those that cannot, or cannot do so easily are understood by all to be the cost of doing business. And absolutely foundational. I see this dark side you mention, June, in experimentation, when teams work so hard to prove the value of their programs that they're twisting and bending to measure and apply the same impact they measured from a controlled environment test with X sample over Z population. But the real value of Analytics programs is providing the company with quality measurement, the ability to get to, use, and interpret the data easily, which also includes data literacy. And so the question becomes - how do you measure the "value" of a culture of data literacy and data informed decision-making where data is both easily accessible and understood?
Simplifying and Automating Marketing Tag QA
4 个月Ultimately, why is about continued and increased investment. Great stuff!
Nice! Any concerns that the goal of 'proving' implies that there is always net positive value at the margin? Maybe it is more 'attention to value' that allows for both positive and negative returns? Keep up the writing!!