#44: The Feature Value Matrix
How to use feature preference and willingness to pay to package products that sell
As a strategy nerd, I love a 2×2 grid.
Nothing forces tradeoffs more than a good matrix. This is especially important when it comes to pricing and packaging for a new product launch.
Because the reality is, not all features are valued equally (nor should they be).
How, and where, you place features within your current packaging structure will not only impact perception of value, but also things like adoption rate, expansion revenue and even conversion rate.
That’s where the feature value matrix comes in. It’s a handy tool to help identify how to package your new products based on two factors: perception of value and willingness to pay.
In today’s edition, we’ll explore this matrix and learn how to leverage it to amplify the success of your product launch.
Let’s get into it 🚀
SPONSORED BY IGNITION
Avoiding the Wasteland
There is a natural tension that exists between every PM and PMM at launch.
The product manager usually wants to release their new feature to the entire customer base, while the product marketer wants to take a more targeted approach.
And while I want as many customers as possible to experience the value of the product, I also know the power of packaging. The feature value matrix provides a tool to have this conversation with your product team without it feeling like you’re taking sides.
Here’s how the matrix works:
1️⃣ Table Stakes - high value, low willingness to pay
These are core features that customers need to extract baseline value out of your product. Table stakes features should usually be included in all packages and tiers.
2️⃣ Differentiators - high value, high willingness to pay
These features are highly desired by both the market and your existing customers. In fact, they’d be willing to pay more to have access to them and they often unlock new value in your product. These highly desired features can be packaged in higher priced tiers to drive expansion opportunities.
3️⃣ Add-Ons - low value, high willingness to pay
These features may seem undesirable or unnecessary to most customers, but for specific segments, they’re a must-have. Add-ons are packaged, unsurprisingly, as add-ons outside of your regular tiers. This way, the folks who want them can pay more to have them.
4️⃣ Wasteland - low value, low willingness to pay
This quadrant should be avoided at all costs. These are features that provide little to no value to customers. When customers see these features in your pricing grid, they may actually view them as detractors. Best to remove entirely.
I like to use the feature value matrix twice during the launch process - upfront while we’re designing the solution, and then pre-release once we have initial feedback from beta customers.
First, I recommend sitting down with your product manager and using the matrix as a tool for discussion. Where would each of you place the feature on the grid? If you’re not aligned, share your perspective and the insights guiding your recommendation.
For smaller feature releases, this exercise is often enough if you’ve gathered the right customer and market data upfront. For larger releases, new product introductions and pricing revamps, I take a more scientific approach.
In this case, I’ll gather two sets of data and then layer them together to place all features on the grid. Here’s the same feature value matrix as before, but with the axes labelled.
First, I run a MaxDiff feature preference survey. This is different than a regular stack rank survey in that instead of asking customers to rank feature value, you instead ask them to identify the most valuable feature and the least valuable feature.
Take the results and calculate the relative preference score for each feature using the following formula: “the # of times the feature is most preferred” minus “the # of times the feature is least preferred” divided by the number of respondents. This will give you a magnitude score between -1 to +1. The closer the feature is to +1, the more desirable it is.
Next, you’ll need to calculate willingness to pay. For this, my go-to approaches are the Van Westendorp or the Gabor-Granger. I’ve spoken about the Van Westendorp in a previous edition if you want to brush up. Once you have the willingness to pay calculated, you’ll also need to calculate a magnitude score like you did with the feature preference.
With this combined information, you can start to place features in the value matrix. For example, a feature with a high relative preference and a low willingness to pay would go in Table Stakes. A feature with a high willingness to pay but low relative preference would go in Add-Ons. And so on.
Want to dig deeper into the framework? Sign up for my new course, Ready for Launch, and I’ll walk you through how to apply this for your future launches!
Introducing Ready For Launch - the PMM's Guide to Product Launches
This 4-week cohort-based course will help you master the art of product launches — from research and strategy, to execution and measurement.
Join Andy McCotter-Bicknell, Jason Oakley, and me as we share the tools, processes, and frameworks we've used to run dozens of successful product launches at companies like ClickUp, Unbounce, Klue, Kajabi, Chili Piper, and more.
The cohort begins October 31 - get the details and claim your spot today.
🎧 Playlist: Queue up my appearance on The Marketing Millennials podcast recorded live from Harvard Business School to get my take on startup marketing strategies