Predicting Saas Success Using The Saas Metrics Canvas
Predicting Saas Success Using The Saas Metrics Canvas
In my previous post: “Saas Metrics Canvas — A Space-Like, Visual, Advanced and Detailed Approach to Track Saas Metrics”, I shared a new visual approach to track all activities and metrics in a Saas company. In this post, I will explain how VCs can use the canvas to predict the future performance of a Saas company and decide whether to invest in a new Saas startup, or to double down on an existing portfolio company, and what returns to expect.
Collecting Data:
VCs need to collect data of ideal values for all metrics of a Saas company. For example, in his “SaaS Metrics 2.0” post, David Skok shared numbers that can set a minimum limit for some metrics, such as:
- LTV:CAC ratio should be >= 3x
- Time to recover CAC should be < 12 months
- The conversion rate from free to paid customers should be 15–20%
- Net Revenue Churn should be < 2% per month
- etc…
There are other important growth numbers to take into consideration such as:
- Neeraj Agrawal’s T2D3 model in his “The SaaS Adventure” post
- Brad Feld’s 40% Rule in his “The Rule of 40% For a Healthy SaaS Company” post
- Leo Faria’s explanation of the 40% Rule in his “The rule of 40% for SaaS and subscription business” post
- and many others
In addition, VCs can take more data from their existing portfolio companies about other metrics. The key here is to take the historical data of those metrics and how much time did it take the companies to enhance their metrics.
Valuation data of successful private and public Saas companies can also be collected. For private companies VCs can use CB Insights, MatterMark, or CrunchBase, or their own data from their own successful Saas companies.
Saas Metrics Canvas Algorithm:
Now VCs need to create their own visual “Saas Metrics Canvas Algorithm”. which is basically the same Saas metrics canvas that their portfolio companies would use, but equipped with a machine learning module that compares all the collected historical data (from above) with the portfolio companies’ data.
With the “Canvas Algorithm”, VCs will be able to “Watch” visually if the portfolio company is moving fast or slow towards achieving unicorn numbers, whats going wrong and where exactly, so fixing any issues can be done faster.
They can also get an estimate of the company’s valuation based on the current metrics compared to other successful Saas companies’ metrics when they were at a similar stage. They can also get an estimate of when to raise a new round, how much is needed, and how much time will it take the company to move to the next level.
VCs will be differentiated by how smarter their different algorithms are, and what reference numbers they consider. Algorithms can become smarter by time and data, so the older and the more detailed historical data a VC can gather and plug into the algorithm the smarter it will be.
VCs can also plug in their own preferences of returns, investment amounts, target exits valuations to identify, in a backwards approach, the target milestones and metrics values over time from now on for their portfolio companies.
Decide On New Investments:
In addition, VCs can use this canvas to decide on new investments. They will simply ask for some input numbers from the startup asking for investment, plug these numbers into the “Canvas Algorithm” and based on a pre-defined standard metrics for different stages (seed, series A, B, C, ….), and other factors of their own, the algorithm can estimate the valuation, and make a suggestion or a recommendation.
VC Algorithms as a Service:
This is what I am actually working on at Algorithm.vc. I am creating a generic algorithm that VCs can use after plugging in their own data and preferences. Although not many VCs believe in this approach right now, but I believe this will be the future sooner than expected!
