Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.
How Data Usage Improves Product Management
Organizations are drowning in data. Yet, product managers may not be using it to the best of their ability. Thinking holistically helps.
5 Min Read
Luis Moreira via Alamy Stock
In today’s fast-paced world of MVPs, CI/CD, and data pipelines, one might think product managers have all the data they need to ensure the success of the products they build. The reality is that some organizations are still suffering from data silos, so it may not be apparent what data is available, let alone its value. Another challenge is a tendency to rely too heavily on technology for everything data related. For example, who has time for focus groups when MVPs and CI/CD pipelines are today’s modus operandi?
AI makes data faster and easier to get, but the insights and recommendations it generates may not necessarily be reliable.
“If you go get data yourself, you tend to pay more attention to it. When it’s just served up, there’s a danger you could get lazy and come up with the wrong conclusions,” says Ramon Chen, chief product officer at data observability platform Acceldata. “All data sources that you use contribute towards your conclusions and product management have to be given a reliability score, so you’ve got to prioritize how you’re going to weight the source of contribution to your conclusions.”
In other words, think strategically about the data sources to be relied upon, though many product managers aren’t approaching the problem that way. Instead, there’s a tendency to just use whatever data is readily available, though it may be incomplete at best.
For example, Acceldata is a SaaS tool, so it’s easy to understand user behavior in a digital sense. Such analysis provides the “what,” but not the “why,” however. So, Chen also gathers input from Acceldata’s product advisory board, talks to the major market research firms, and uses customer feedback data. More recently, artificial intelligence was added to the mix, and it yielded some interesting results.
“We added a new AI copilot capability to our data observability platform and that was born out of data gathered in a normal way,” says Chen. “Now, AI is helping us develop an AI add-on and it’s generating revenue. We would not have necessarily built this had we not been able to synthesize that data.”
Data-Driven Product Management Can Drive Revenue
Aislinn Wright, VP of product management at Postgres data and AI company EnterpriseDB, has a similar experience. Her company had a lot of customer data in Salesforce, but as a product manager, she wondered why deals were won and lost. So, she pulled the data from Salesforce into Tableau and conducted some analysis related to pricing.
Looking at data from all different stages, Wright discovered pricing was a factor in lost sales, but the analysis also surfaced a new opportunity: alternative pricing.
“We added the option for customers to purchase instance-based pricing, because some of the core price points are too expensive for customers that are spinning up larger footprints. We tested a full open-source plan with instances when previously we had none of that, so it gave us the opportunity to market to a whole [new] segment of folks,” says Wright. “Had I not gone through some of the analysis on the Salesforce side in terms of the loss reasons, we probably never would have come up with this new plan, offering instances and additional cores and changing some of the plant structure.”
Wright and her team also changed the way data was collected from the sales team, so she could capture some more specific loss reasons related to products, which helps with the roadmap.
For the past year, Wright has also been working with a user research firm that ran 15 different personas through a one-hour prototyping demo with the goal of getting product feedback.
“That worked really well,” says Wright. “Those four projects [provided] more detail about some of the roadmap changes, which resonated really well from a messaging standpoint and understanding what [customers] wanted to see in the next version.”
Think Holistically
It’s wise to think about products at a holistic level, from a full customer experience point of view. That requires both qualitative and quantitative data.
Jeff Piazza, SVP of experience design at digital transformation and product development services firm Orion Innovation, says his company starts with primary research to inform both the design team and the product management team in terms of what features to include or not include in an app.
For example, in the case of a contractor app, they worked hand-in-glove with that client on market research. As part of that, they studied the market, voice of the customer, and more.
“We’d send out a couple examples of a flow and get an understanding of whether it would resonate with the audience or not. In addition to that, we spent time at their partnership events. We set up an innovation booth that allowed us to be on the ground and prototype using Figma. [We walked] through the different scenarios to really understand how people would use this day to day,” says Piazza. “That allowed us to understand what to include or not include before we move into a build phase.”
When conducting such research, it’s important to ask the right questions. One common mistake is to fall victim to confirmation bias, which involves asking questions in a way that confirms the researcher’s assumptions when the most valuable research reduces such bias. A better approach is to use the scientific method and test a hypothesis.
“With us, it’s starts with the question, ‘Are we asking the right question in order to ensure that we’re unbiased with the result?” says Piazza. “[When it comes to research,] we think about not only the overall span of building the application and the scope of the application, but also how we’re going to market.”
That involves understanding what the app issues are and what the current experience of a contractor is, such as how long it takes to complete tasks (since the goal of the app is to shorten the time it takes to perform those tasks). Understanding the time it takes to complete a task without the app serves as the baseline for benchmarking so it’s clear how much time the app saves contractors.
“Start with the idea of understanding the ideal experience, not just from a traffic and metrics standpoint -- OKRs and KPIs -- but also from an emotional standpoint because emotion is what is going to drive people from like to love,” says Piazza. “Understand what it means for you and your target market and that particular product and work backwards to understand how you get there.”
About the Author
You May Also Like
More Insights
