Every time we start a Product Cycle for a new product, one of the points that we want to be defined from the right beginning is its KPIs (Key Performance Indicators).
There is a huge variety of possible KPIs for digital products, as for example:
- DAU/MAU (Daily/Monthly Active Users)
- Customer Lifetime Value (CLV)
- Customer Acquisition Cost (CAC)
- Net Promoter Score (NPS)
- Number of downloads
The natural tendency is for Product Managers to pick a big handful of KPIs that they keep improving over time during the Product’s development.
But there are two problems with this: 1 - Teams will be pulled in many different directions, making more harm than good. i.e., they will be trying to optimize towards many different aspects, which will end resulting in very low progress or eventually none. 2 - This is a sign that teams don’t really know what they’re product should be doing better than any else.
Check out a few good practices when choosing your main KPIs!
Main KPIs focus on delivering customer value
KPIs should sit around what you want to achieve with your product. And while at first thought that may sit around financial metrics as revenue or number of users, teams should be aiming to the value that the product should be delivering, and making sure that they use it enough so that they come back to use it more.
The classic DAU/MAU are often vanity metrics - they may give a signal of the amount of traffic but they are not an indicator at all for whether a product is really working. If it’s achieving its purpose.
You should focus on what users were doing and why they were doing it. So for example with Soundy - Giphy for sounds, - instead of focusing on DAU/MAU we focus on “Number of Sound Plays” and “Number of Uploads” - that’s the value we are providing with the product, so that’s what we should focus on.
And since every product has a different purpose, so should their main KPIs. Your product and business are unique, so don’t try to paste another product’s metrics to it, and work on your own instead.
(…) teams should be aiming to the value that the product should be delivering and making sure that they use it enough so that they come back to use it more.
Teams should relate to their main KPIs
It’s very easy to measure performance using big picture goals, but it’s common for them to not be relatable by some of the teams working on the product. It’s common for product teams to comment that having a single big metric for every team to not work properly because most individuals couldn’t link their direct day-to-day work to that one metric.
For these cases you should be looking for cause and effect KPIs - KPIs where there are a causal relationship between the action the statistic measures and the desired higher level outcomes.
So for example, if your product’s main metric is “Number of minutes played”, the marketing team maybe should focus in the “Number of signups”, which should (in theory at least) cause a higher number of minutes played.
(…) you should be looking for cause and effect KPIs
Make KPIs easy to get
Imagine the case that your main KPI is Customer Acquisition Cost (CAC). To get an accurate reading of CAC you probably need to aggregate data coming from a range of sources, from marketing, sales, and possibly even development, with origin in tools like spreadsheets, the company’s CRM, Twitter Ads platform, and maybe the time tracking or Project Management tool used by your dev team. This means that to make this laborous CAC easy to get you should automate it’s calculation at least until some degree, or have someone responsible to manually calculating it on a frequent basis (ex: daily).
Otherwise, it will be too difficult to get this important KPI, or you’ll be prone to make errors when calculating it, and thus lose the purpose.
You should ask: Do these reasons still hold true? Has our business or the context within which it operates changed? Can our metrics be refined to suit these changes?
Review your KPIs often
Take the time to review the reasons for selecting specific KPIs on a periodic basis. You should ask: Do these reasons still hold true? Has our business or the context within which it operates changed? Can our metrics be refined to suit these changes?
Also, ask if there’s a real cause-effect between the chosen KPIs and the desired outcomes. Often, people’s deep confidence in their judgments and abilities is often at odds with reality. Maybe your confidence in setting “Number of signups” as the main goal is actually not translating into higher “Number of minutes played”, as you can now see through the acquired data.
You need to make sure you’re choosing the key supporting metrics that correlate with the higher frequency of use of your product’s features and how likely they will stay using it vs churning out.