6 ways to integrate a data strategy into your product workflow
Mar 21, 2018
Even though most product and engineering organizations today are collecting tons of data, most only use a fraction of what is available. This is often due to competing priorities or lack of implementation experience in data strategy.
Looker and Atlassian are working together to make data-driven product development workflows a reality.
There’s no silver bullet when it comes to making data-backed decisions the norm, but there are steps you can take to shift the culture of your team, or even your company.
1: Expand your definition of data
Data from instrumented products is increasingly valuable to product teams because it shows how users interact with the product directly. But data from the product itself isn’t the only source of information you can use to improve your product and workstreams.
Data from issue tracking tools can provide valuable insight into historical development performance and can help you plan for future development cycles. Our ETL partners, like Fivetran, make it easy to move data from applications you use to run your development team.
Check out the new JIRA by Fivetran block that makes it easy to get up and running with Jira analysis.
2: Broaden your focus with data from other departments
The most successful product analytics often incorporates a wide range of sources outside of traditional product data to uncover insights into areas of the product effectiveness and broader customer experience.
Marketing, finance, and sales data can provide valuable insight to a product team. Not only does this new data provide a fuller picture of how the product is received, purchased, and used, but it can also determine the larger impact of product development choices. For example, you can see the impact of a new feature on revenue, the impact of a bug on customer retention or the impact of a marketing campaign on the types of new users.
Services like Fivetran make data consolidation with Jira easy and scalable.
Your data should be representative of your company as a whole, not just your department or function because at the end of the day, that is how your customer experiences your product and brand. By looking at the full picture, you can get a much richer and more accurate understanding of your product and your users.
3: Reveal your metrics that matter
Once you have the data you need, keep your product team focused on the task of building great products by developing clear, team-wide definitions of key performance indicators you can all agree upon. This includes not only product performance but customer engagement, as well.
Find your metrics that matter, then break those larger metrics down into levers that each individual on the team can begin to influence. Monthly active users is a great North Star metric but is difficult for an individual to influence directly. Increasing usage of a sticky feature, on the other hand, is a measurable goal an individual can work towards.
4: Apply potent metrics to every conversation
Once you have actionable metrics, make them a conscious priority for your team. Talk about these metrics in meetings and bring them to the forefront of group conversations.
In addition to in-person conversations, your product team likely uses a wide range of tools in your day-to-day workflows. Email, chat, and project management software all play a role in how we coordinate, track, and communicate. Set up alerts in these tools so the whole team knows when certain thresholds are passed or when unexpected events occur so you can act quickly.
Just having metrics isn’t enough. To make truly data-driven decisions, metrics need to be routinely baked into every conversation.
5: Shrink the distance between insight and action
Data ensures you’re informed when making decisions, but the larger the distance between the information and the action, the harder it is to keep the two connected.
Ideally, this distance would be almost invisible. If you have product issues to file, make sure you can do so right from your data tools without delay. If you are describing an insight in a Confluence page, insert that visualization directly so the reader can see the live data in context.
6: Treat data analytics development as if it were your own product
Data is only powerful when metrics remain relevant. Which is why it’s very important to enable reporting on a single version of the truth.
Your data tool must change and grow with your business, and to do this, you can pull best practices from other development processes you already know and employ. Rely on sprint cycles to build new analytics tools and use version control to keep improve analytic outcomes, encourage collaboration and pivot data development quickly when necessary.