Customer Validation

Optimise your Value Proposition with Card Sort

How to get a read on what features customers find important

Erik van der Pluijm
Erik van der Pluijm
Nov 08 2019 - NaN min read

It’s all in holding the right cards — Photo by Daniel Prado on Unsplash

You've created a Value Proposition that you want to test. Either because of uncertainty, creativity, or feature creep, you've got way too many options. How can you get your customer to help you find the most important features?

Validating a complex set of features

Let’s say you are validating a new product, and you’re homing in on a value proposition. So far so good.

However, it turns out you’ve come up with way too many ideas for features based on your customer’s input.

How to make sense of this? How to whittle this long list of features down to a manageable MVP?

And, it’s not just selecting a few of the most interesting features: the total mix of features will influence the desirability of the MVP for your customer. Clearly, validating feature combinations one by one will take way too much time.

There has to be a way to get customers’ reactions on combinations of features fast, right?

Card Sorting

Luckily, there is! A simple tool can help you do just that: the ‘card sort’. This technique originates in usability research and UX design, but it is a great trick to use in validation as well.The idea is that while it is difficult for people to prioritise a large number of (possibly conflicting) options mentally, they are very capable of prioritising these items when they can visually organise them in an ordered list.To do that, they get a bunch of cards, each labeled with a single feature or item, and are asked to order these according to some organising principle.For instance, order a set of MVP features by putting the most desirable features on the top, and the least desirable ones on the bottom.

Tip! More information on how to do different types of card sorting can be found here.

Example: Card sort for problem validation

For the project ‘Square One’, an accelerator sprint for early stage startups, I created a card sort exercise that allowed us to facilitate validation meetings with potential customers, and to shape the value proposition of the Startup Scan together with them.I created a total of 32 cards, printed on A6 (‘postcard’) size. These were color coded in different categories, as I wanted to understand what the tools, topics, questions, and goals were they found most interesting. The color coding allowed me to easily find the appropriate sets of cards to use in a given situation.Some examples of cards:

One of the cards used in validating the most important topics for startups / corporate ventures

One of the cards used in validating the most important topics for startups / corporate ventures

One of the cards used in validating the most important topics for startups / corporate ventures

One of the cards used in validating the most important topics for startups / corporate ventures

Having conversations with startups and corporate ventures using these cards had an extra benefit. Besides getting information on their priorities and their problems, they also worked as conversation starters on topics that were top of mind for the potential customer.

Example: Card sort for MVP definition

This week, I met over coffee with innovation coach Josse Dorleijn from Moonshot Missions, who was one of the first alumni of the Experiment Cookbook course.

He told me he just finished a project where he helped transform a team coaching agency from a purely consultancy and coaching business model towards a more scalable self-service business model, by productising their offering. In this project, he made great use of card sorting to get feedback on different value propositions and MVP candidates.

He has a few tips when using Card Sorting:

  1. Keep asking ‘why’. Why do they prefer option A over option B? Why do they make a certain choice? Asking this will get you really valuable information you can use to shape your MVP.
  2. Add a few empty cards and ask what features customers miss.
  3. Have the customer organise features in ‘must haves’ and ‘nice to haves’.

Tip! If you want to read more about Josse’s approach, check out this post on LinkedIn — it’s in Dutch, but Google Translate is your friend* 😎

Tips and tricks

First of all, card sorting is not limited to text.That makes it a really powerful tool when you want to know more about things like style, emotion, and the feelings customers have. You can show them different types of photography, or logos, and have them sort these, for instance. You’ll be surprised what they have to say when you ask them about their preferences.

Second, when you plan to do a card sort, make sure you have at least 10–15 options.

And lastly, although it is definitely possible to have people do a card sort with post-its or handwritten cards, it is worthwhile to spend a bit of effort in making them look ‘official’. Printing a set of cards (as in the example above) does not have to be expensive, and you can keep re-using them for different interviews — and it means your customer will take the exercise much more seriously.

Next steps

The feedback from your customers on the priorities of features gives you an idea of the value they assign to these features. This can help you come up with hypotheses of combinations of these features, for instance to create pricing plans for your MVP.

However, you can’t trust customers to come up with packages that make sense all on their own. They might combine features that are interesting for them, but require a lot of work for you to realise as a combined MVP. They also might mix and match between fundamentally different approaches.

Therefore, it is time to validate their input.

Take the feature combinations they have suggested, and shape them into 3–4 different flavours of MVP. Use your knowledge of the constraints that shape the feasibility of these feature combinations to inform these MVP flavours. Then, run a new experiment where you show these MVPs and try to get traction from your customers.

  • Do they find them acceptable?
  • Do they want to sign up or buy?
  • Do the differences in response to the different versions agree with what you expected?
  • This will help you come up with great MVP candidates quickly and effectively.

Keep experimenting 🚀

Read this post on Medium