A key challenge of working with KPIs is to select the right indicators: There are so many different metrics to choose from including daily active users, net promoter score, and profit, to name just a few. What’s more, senior managers and stakeholders can have strong views on which indicators should be used. This article helps you select the metrics that really matter and are truly helpful for your product.
An audio version of this article is available at: https://www.romanpichler.com/romans-podcasts/how-to-choose-the-right-kpis-for-your-product/
What are KPIs?
Key performance indicators (KPIs) are metrics that measure how your product is doing. Effective KPIs help you understand if your product is creating the desired value for the users, the customers, and the business. Without KPIs, you end up guessing how well your product is performing. It’s like driving a car with your vision blurred: You can’t see if you are heading in the right direction or getting closer to your destination. You might have a hunch, but you don’t know if it is correct. Using KPIs and collecting the relevant data helps you balance intuition with empirical evidence. This increases the chances of making the right decisions and achieving product success.
A Goal-directed Approach to Choosing KPIs
To select the right KPIs, I recommend taking the following three steps: First, use the user and business goals in the product strategy to select an initial set of indicators. Then take into account the product goals on the product roadmap to discover additional KPIs. Finally, choose further indicators to assess the health of your product and team. This, of course, assumes that you have a validated product strategy and a realistic product roadmap in place. If that’s not the case, then I recommend creating those two plans first.
As you may have noticed, I suggest using a goal-directed approach to identify the right indicators. I am not a big fan of “standard” KPIs, for example, customer acquisition cost (CAC), churn, and number of active users for SaaS products. While it’s helpful to be aware of the indicators that may be commonly used for the type of product you manage, it would be a mistake to blindly adopt them. Employing user, business, and product goals to guide the selection of KPIs avoids this mistake and it ensures that your indicators are relevant and helpful.
Step 1: Take Advantage of the Product’s User and Business Goals
To select the right KPIs, start with the needs and the business goals in the product strategy. Then consider how you can tell if they have been met.
Say I want to offer a product that reduces the blood sugar levels of individuals who have type-2 diabetes and that generates £50k of revenue within the first 12 months of launching the app. I would then ask myself how I can measure that the product creates the desired user and business value. To determine the former, I might choose user feedback, net promoter score (NPS), and referral rate as the key performance indicators. To understand the latter, I might select monthly recurring revenue (MRR).
Note that this approach assumes that valid user and business goals are available. In other words, the goals should be part of a validated product strategy — a strategy whose key assumptions and risks have been successfully addressed. This might have involved observing target users, interviewing them, carrying out competitor research, and using throw-away prototypes, to name just a few strategy validation techniques.
Step 2: Use Product Goals to Discover Additional KPIs
In addition to deriving KPIs from the user and business goals, I like to use the product goals I have captured on the product roadmap in order to discover additional indicators. A product goal is a specific and measurable benefit or outcome your product should achieve, typically in the next two to three months. What’s more, every product goal should be aligned with the user and business goals in the product strategy, as I explain in more detail in my article “Product Goals in Scrum.”
Say the goal of my initial product (MVP) is to allow the users to better understand their eating habits and acquire an initial user base. I would then look at the first part of this goal and ask myself if I need to introduce a new indicator. As I’ve already chosen user feedback, net promoter score, and referral rate, I would not add another metric at this stage. But the second part, acquire an initial user base, would require the introduction of a new KPI in order to understand if the acquisition goal has been met. To do so, I might measure market share, for instance, by tracking the product’s position in the appropriate app store.
Step 3: Add Health Indicators
Measuring how well your product is doing at meeting its user, business, and product goals is great. But it is not enough. Let’s say that my new product is meeting its revenue targets, that the user feedback is excellent, and that NPS and referral rates are high. This suggests that it is performing well, and there seems to be no reason to worry.
But if the team motivation is declining — as people are being overworked or stressed — or if the code quality is deteriorating — as bugs are accumulating and the code complexity is increasing, alarm bells should go off. These indicators suggest that achieving product success will be much harder in the future, due to an increase in technical debt, higher absenteeism or turnover rate.
It is therefore helpful to add metrics that measure how healthy your product and team are and that allow you to spot important warning signs early on so you can be proactive and respond to them in a timely manner — instead of reacting to them with the back against the wall. These indicators include product quality, schedule variances, which is the ability to meet product goals on time and budget, team motivation, team knowledge and skills, and stakeholder engagement.
Less is More
Having been involved in product management for 20 years, I sometimes marvel at how much our profession has changed for the better. This includes the analytics tools we have available today. These powerful tools usually generate more data for us than we can handle. And herein lies the issue: There is no point in measuring everything that can be measured.
Using more KPIs than necessary and analysing data that provides no value is a waste of time. In the worst case, you act on irrelevant data and make wrong decisions. Think of driving a car. A small number of indicators are helpful for safely getting to your destination, including how fast you are travelling and how much battery/fuel is left. If the car dashboard continuously showed a range of other data, such as tyre pressure or ABS status, it would be harder to take in the relevant information and make the right decisions while driving.
Therefore, limit the number of KPIs you use. Carefully select them by using the factors discussed above, and regularly review and adjust them. Don’t make the mistake of allowing powerful stakeholders to dictate KPIs to you. Instead, explain to them how effective product KPIs are derived, and consider inviting them to the next product strategy workshop in order to collaboratively determine the right KPIs.