How data-informed UX helps avoid Minimum Viable Product bloat

The most important question designers need to ask is what value are we providing to users?

Designers play a more significant role in creating a good Minimal Viable Product (MVP) than you might think.

Part of becoming a Product Designer is re-learning terms I thought I knew. MVP is one such term: in product terms, it’s a minimal set of features you build to attract customers and validate a product idea.

However, it’s only recently taken on a new context to me upon reading Hacking Product Design, by Tony Jing: MVP, as he defines it, is about offering minimal value to the target audience with your product that they’re willing to become users.

Looking at it that way provides a clue into the primary question designers must always consider: does your product offer enough value to your users? You must always keep that in mind if your team asks you for any last-minute changes.

Scaling down designs or designing by Engineering

I’ve found, especially on teams who build out Minimal Viable Product (MVP) quickly, that it can become Design by Engineering at a certain point.

The Product has determined the problem and a set of features, Designers have built mockups and prototypes, and then it’s up to Engineers to build a specific set of features in a timeframe. This is where Engineers can point out technical issues or other things that weren’t noticed, and some tough design decisions could be made.

For example, Engineers could point out that showing a Data Visualization of all aggregated traffic on a page would take 15 minutes to load. In this case, teams will be forced to scale down interactions, cut features, and more to get the MVP out on time.

As a side note, this is also how UX debt accumulates.

However, Designing by Engineer is just as dangerous as designing by any other stakeholder. The reason is simple: if you build something easy for your Engineers to build, but your users don’t want to use it, there’s the chance that the product will get scrapped (or pivot away), wasting a whole bunch of time and effort.

This is where it’s our job, as Designers, to keep our minds on one question: what’s the minimum value I bring to a user? As long as that exists, in some scaled-down form, you can continue making iterations towards things that users love.

But how exactly can you keep track of this ‘minimum value’? By using a hypothesis and Data-Informed UX Design.

The Data-Informed UX Design view makes MVPs and minimal value easy to define.

This is because Data-Informed UX Design is about defining problems and actions in terms of hypotheses. MVP is about testing such hypotheses and validating whether a product solves a particular problem for a user.

If the concept is unfamiliar, let’s discuss how you usually iterate on designs. First, you would likely gather user feedback about the design and workflow, which you would then use to inform the next iteration.

Data-Informed UX Design adds two things to this process:

1. A way to quantify success or failure (through metrics)
2. A structure to define the value to the user through designs, user actions, and metrics (with hypotheses)

For example, imagine that you’re designing a website that focuses on building a community with user-generated content like brainstorming templates (such as Miro). The primary metric we care about is Monthly Active Users (MAU) because we want an engaged community.

When you do research, you find that users will get value from browsing community templates and downloading/using them, saving them the time to create their own from scratch. You could then come up with the following hypothesis for MVP:

“Users struggle to find templates they can use for workshopping, which often results in them having to create their own (or search the internet for them). By creating on an account on our website, they can download, customize, and start building in the templates right away instead of struggling to find templates. We will know we are successful at this when our monthly active users increases by 15%.”

Having this hypothesis in mind will help you when it comes to an understanding if Product or Engineering decisions are going to make the MVP worse for users.

For example, if Product wanted to force users to sign up right away and wait for an e-mail link before actually building in a template, you could push back, stating that part of the value we offer users is having them use our templates right away.

Likewise, if Engineering says we can’t have users save multiple templates to an account, you could still work to ensure they can save and quickly use one template.

Doing this ensures that even if things need to change for MVP, they still support the user in how the product is designed.

Minimum Viable Product (MVP) means Minimum Value to Users

In your day-to-day actions, it can be challenging to always keep the big picture in mind, especially when it comes to concepts like MVP.

However, the one thing that we always need to remember, as Designers, is that the products we design must be of value to our users. Otherwise, we’re building solutions to problems that don’t exist.

While you can have some idea of the value you’re bringing to the users while designing something, last-minute changes to the MVP could catch you off-guard and ultimately make it worse for users.

So formulating that hypothesis ahead of time, through Data-Informed UX Design, can ensure that you can always measure the decisions the team is advocating for against the value you bring to users.

Doing so will ensure that what’s valuable to users doesn’t get confused when your team is building it.

Kai Wong is a Senior Product Designer, Data-Informed Design Author, and Data and Design newsletter author. His new free book, The Resilient UX Professional, provides real-world advice to get your first UX job and advance your UX career.