There’s no getting around it - product analytics matters. As we’ve already explored, intelligent use of data as an integral part of the product development process is often the differentiator between successful and unsuccessful products. Unfortunately, putting data at the heart of the product development process is hard and the obstacles are not always what you might think.
We’ve been privileged to work with some brilliant product development teams across our customer base. In this post we’ll outline how companies that successfully use data in the product development process do so. Hopefully, this will prove useful for product teams that want to be more data-driven by providing a clear picture of what successful use of data in the product development process looks like in practice. Though technology does have a role to play, it’s a means to an end and process and culture are much more important than technology for reaching that end.
These insights were not formed in a vacuum but come from perspective based on our experience with our own clients, and countless conversations with product managers across different industries and geographies. But this is an ongoing conversation and we’d love to hear from across the product analytics community whether their experience chimes with ours, and particularly if they’ve seen differences. Comments and counter opinions are very welcome in the accompanying discourse thread.
Product development is an iterative process, organic and consistently flowing. Product teams aim to understand their users and the role the product plays in their users’ lives: this might be measured in value (e.g. for a trading application that is literally being used to generate profit for traders), time spent being entertained or informed (in the case of media or games companies, for example), or some other measure tied closely to the purpose of the product. As their users’ needs change and evolve, sometimes in drastic new directions, product teams are constantly looking for opportunities to improve their product so that it better serves existing use cases or can start to serve new, potentially adjacent use cases. Early on in the product lifecycle, product teams focus on finding product-market fit. Once that fit is established the focus switches to improving the user experience, driving more value, for users or the business or both. Updates are proposed, prioritised, developed and launched and rolled out:
Data has an important role to play in a number of the stages identified above.
There is a wide misconception that data is reactive and is only useful for looking at the past. The truth is that data can play a significant role in getting product teams to be forward thinking by helping identify opportunities to improve a product. Quantitative data describing how users actually engage with products might be used to uncover if the product is being used in a way that hasn’t been anticipated, or if particular workflows are cumbersome and awkward. Qualitative feedback can be used to identify how users feel about different aspects of the product, potentially highlighting areas that frustrate users, even if the quantitative data suggests that users still successfully navigate those parts of the product.
Product teams that are good at data use it as part of the creative process for identifying opportunities to improve a product by using it to highlight parts of the user experience that are suboptimal.
When it comes to specifying updates to products, we see a big difference between those product teams that are good at using data and those that are not. When an update is specified to a product, data-sophisticated product teams will include, as part of the specification:
The above two steps are essential for any product development teams that want to be data driven and they can be as challenging as they are useful. Unfortunately, the majority of product teams do not go through this rigorous exercise - limiting their ability to use data to effectively inform the product development process.
As part of the regular sprint planning process, product managers will be forced to choose how to prioritize different potential updates to the product. This decision can be informed by the clear, quantitative benefit that the team hopes to derive from the update, specified as part of the overall specification process.
Good data-driven product development teams will have a formal process for A/B testing any new features with a subset of users prior to any rollout of that feature. Remember that the product team will, as part of specifying an update, already have specified the impact that that update should have. A/B testing provides a formal mechanism for measuring if the update has that desired effect.
With each new release of a product, therefore, a new set of A/B tests will be launched to test the latest batch of features. At the end of the test period the results of those tests will be assessed. Only then will product teams that are serious about data roll out those features where the A/B test showed that the update had the desired effect to the rest of the user base. Otherwise the feature will be canned.
Having the discipline to stick to this formal process is the only way product teams can ensure that over time successive releases drive significant incremental improvements to the products.
A comprehensive look at the tools, technology, and culture companies need to effectively use data
As should be clear from the description above, data-driven product teams follow a very formal, detail-oriented process, where data plays a role in:
This is not easy: it requires a huge amount of discipline from the product team. It requires additional overhead in terms of upfront work in specifying the update. It means that every update has to be A/B tested, which itself has an overhead. And it means product managers have to be open to canning features they may have strong emotional attachments to, because the data tells them otherwise. The return on that extra effort and discipline however is enormous, especially when measured over time. It drives a deep understanding of how user behaviour shifts with each product update across all members of the product team, and means product teams can be very confident that over time they will drive systematic improvements in the performance of the product.
In practice, adherence to this strict process for managing the role of data in the product development process is only possible given the right company culture, one where:
Organisations differ in the extent to which they value both data and experimentation. Too many organisations only pay lip service to the value of data. (Data is quoted in arguments, but often is not decisive.) Too many organisations are not committed to experimentation. Committing the resources necessary to run all the different experiments a process like this requires necessitates a very high level of organizational commitment. Only those organisations that really value the solid insight that derives from that experimentation will make that commitment. For those organisations that do commit the resources, that commitment, to both data and experimentation, will be an important part of the organization’s culture.
Product teams in organisations that do not have that deep cultural commitment to data and experimentation will constantly be under pressure to cut corners, making it very hard for those team members, however committed they are to data and experimentation, to do data right.
Technology is important: as we’ll describe in a later post, it plays an essential enabling role in making this process possible and not unnecessarily onerous. However, the value that technology can unlock will be contingent on the product team, and the organization that operates it, having the right culture and process. Without those, the best new product analytics solution in the world will not help that product team use data to drive that competitive advantage. A data-driven product development process is the only way to understand your users deeply enough to build the best product for them, and if you don’t, someone else will.
To learn more about product development with us, make sure you subscribe to our newsletter for all of the latest entries in our product analytics series.