Prioritize product features by reducing uncertainty
It exists many solutions to prioritize features, like RICE or MoSCoW.
The VTU (Value by time and uncertainty) allows to discover problems, reduce uncertainty and compare features on the same level.
Enjoy the google template!
The problem cost by customer ($)
In order to compare features on the same level, the most difficult challenge is to estimate the problem cost. It must be based on a similar period for each feature (month by example).
Let see an example based on this problem: “Missed appointments for our persona: Doctors, aka No show”.
By using data or a survey, you can get the average problem cost: “The unbilled value($) of all appointments missed”.
The market problem cost ($)
The second step is to calculate the global problem cost on the market.
It should be based on the SOM (Serviceable Obtainable Market).
Sometimes, it’s useful to get the problem cost by persona and measure it*.
The market value proposition($)
The next step is to estimate the impact of the solution. Actually, how many problem percentages will be reduced?
Several methods exist:
- get results from customers that are already using this type of solution.
- organize a manual test: for the current example, it’s possible to send SMS manually. (Do think that don’t scale)
- prototype a quick solution for beta testers
With our example, it appears that the “SMS reminder solution” reduces missed appointments by 60%.
The market solution value is : 85,000,000$ * 60% = 51,000,000$.
In other words, if it was possible to set this solution for everyone, it should generate 51,000,000$ of new revenue.
The value by time ($)
It’s necessary to evaluate the time needed to implement this solution. It could be story points. With your example, it needs 100 days.
The value by time (VT) is : 51,000,000$/100 = 510,000$.
Every working day will provide 510,000$ of value to the market.
The value by time & uncertainty($)
Optionally, it’s useful to introduce a level of uncertainty. The template allows setting an uncertainty from 0 to 10 for each item.
For our SMS example, the uncertainty is 4. Only 60% of the VT should be considered.
VTU = 510,000$*60% = 306,000$.
Compare results
At the end, it’s easy to compare features :
Final considerations
It‘s difficult to calculate the problem cost but, at the end of the day, customers will compare a value proposition with his cost.
This framework is also useful to sell the product :
- Understand the problem, be able to calculate the cost for each customer, will be useful on all the AARRR process.
- Understand the problem cost by persona is fundamental. Indeed, the solution value by persona should be compared to the Customer Acquisition Cost. It’s the key to success.
This framework should not be used when:
- The time to build is not significant. But be careful to consider also the product debt that will be generated. (build, support, complexity to update etc..)
- The problem cost is too complex to estimate. But it’s important to try!
- The budget allowed to discovering is not enough important
My advice, determine the budget and the time you have to discover and do your best to reduce uncertainty using this time.
Innovations come from an important problem and an efficient solution. It should be a priority for every tech company to afford a significant budget on discovery.
*All data on this article are purely fictive
