Samaipata’s cohort analysis 2.0
Break free from your tech team, become a cohort expert
Welcome to Samaipata!
We are an early stage founders’ fund investing in digital platforms, displaying increasing returns at scale, across Europe.
If you are a pre-seed or seed stage founder, you probably still haven’t got in your hands neither a data warehouse nor powerful visualisation tools for the different analytics, therefore this post is for you.
Cohort Analysis is one of the most important techniques in order to fully understand the behaviour of a certain part of the population. In marketplace dynamics, understanding the behaviour of your users and being able to see both sides of the equation (demand and supply) are some of the most important tasks… especially when you are trying to build the next multi billion dollar business! Therefore, we have tried to build a spreadsheet where you can, without any previous knowledge or understanding of what a cohort is, break free from your tech team. The key takeaway is for you to not bother them every time you need a query or want to run a retention analysis, and become yourself a cohort expert in a matter of minutes.
Yes, up to here it all sounds very promising, however, the million-dollar question is: how does this actually work?
Cohort Analysis as a base to understand unbiased data
The main aim of cohort analysis is to eliminate as much bias as possible from your whole database in order to further understand the different trends within your business. To do so, the first thing you have to do is a query from your database in order to get all the information you need. Full stop, the beauty of this spreadsheet is that from this point onwards, you will not require the help of your tech team.
Once you have the whole information, you can jump into the model we’ve built which explains the train of thought that goes from the retention and engagement of your startup to how much revenue will a first time user generate to the business during his lifetime. This is what we will call the Lifetime Value. Once we have final target defined, understanding the process will allow us to be a kick-ass data-driven person with solid arguments to support strategic decisions.
Just to be clear, some quick basic concepts that I think are worth recalling before getting started are:
- Cohort: in statistics, a cohort is a group of people who share a common characteristic. (e.g. a group of customers from the same city, who are female, who were born the same year, etc). Thus, a cohort analysis is the study of how a certain group of people who share a common characteristic behave across time.
- Events: an event is an action made by a user in the platform and it is used to measure the engagement that users have to your business. In social media cases it could be commenting or uploading a picture while in a transactional marketplace it could be a purchasing event.
- First Time Users (FTU): it has become a standard to define cohorts as the group of people who became FTUs in a given date (usually a given month, sometimes a week). By doing this, we are able to estimate a median behaviour across the different groups of people that joined in radically different periods of time where different decisions were being made: vouchers, promos, etc.
Quick walk through the model
If this still sounds a bit unfamiliar, let me take you step by step through this journey. I know this might sound like the least intuitive thing on earth, but it will be damn useful, trust us!
I believe one of the most usual questions that come to the top of our mind when we see a new tech start up is: What is the quality of their user base? Well, we consider this has a double edge.
- Retention. As most of you have already guessed, it shows the percentage of FTUs who continue interacting with your platform in their second, third, etc months after their first arrival to the platform. It is very important to understand this since it can be misleading.
- Engagement. On the other hand, we find that engagement should not only be measured by retention but by the number of events that your users are generating in the business in their recurrent months. This makes sense, you can have many users still interacting every month, however, if you are hoping to create a long-lasting business, the idea is that as the months go by your user base is more and more engaged with your platform. I guess the ideal scenario would be for your users to feel like they “need” you more and more as time passes.
The ideal user scenario
Once you got those two things sorted out. Bualá! You magically multiply them and you get one of the most important conclusions out of this analysis. We have called this “RxE” and it representes the average events per FTU. I know what you are thinking, we are the most creative people on planet earth. This metric is measuring the number of events, on average, that a returning FTU has generated during his time at the platform. A good quick read to this graph can be: if I gain 100 users tomorrow, how many events are they estimated to generate per user in their upcoming months.
Economic value is the light at the end of the road
Okey, we are almost there. Up until now, you have been focusing in a non economic value of your platform. If you have a transactional business, once you have your RxE (Average events per FTU) calculated, by multiplying it by the Average order value, you will be able to translate interactions to economic value. Here, you have now found your way to a stage where you have the Average Estimated Revenue per FTU. This answers to the question: “How much revenue can I expect from my FTUs in their x month in my user base?”
Phiuuuu! We are almost there. Once you got this, we can now present our final trick. This is called the Lifetime Value. Basically, all you have to do is to calculate the accumulated graph of the Estimated Revenue per FTU and you will get a positive slope line indicating the accumulated revenue that a FTU will have generated to you across their lifetime. Therefore, in the last month, you will find the key data point, that is the amount of economic value a FTU will have generated to your business by the time they leave.
Download the template and get your hands dirty!
Okayyyyy! All this being said, I promise we have tried to make it as easy as an analysis of this kind can be so that you can dedicate your time to fun stuff. The template you can download here is supposed to be intuitive and you’ll just need 5 minutes to complete 3 steps in order to build your first cohort analysis. Stop bothering your tech team with things you can do on your own and let them focus on the hard stuff.
What are you waiting for to try it out? If you have any further doubts with interpreting or updating the template, we’ll be delighted to help you. Email us! If you have any problems with the download please contact us. We have tried to make the file as light as possible but sorry to say, we haven’t come up with that magic trick just yet. Hasta la vista!
And as always, if you’re a European platform founder looking for Seed funding, please send us your deck here or subscribe to our Monthly Founders Kit here!
