Building a Minimum Viable Product – No Time For Edge Cases

Edge case

Does this look like an edge case to you?

Like many startups, at Unbounce we’re often faced with the dilemma of which features should be in, and which should be out. Particularly for the most crucial of releases – product launch.

We’re taking an approach known as the MVP (Minimum Viable Product) for our first release. This refers to the absolute minimal set of features and functionality required to produce software that can be taken to market in a state acceptable to it’s early adopters.

This requires debate on each functional aspect of the Unbounce landing page service. It’s taught us to exercise considerable restraint, scrutinizing the value and the “is it absolutely essential” factor of each requirement.

Feature In/Feature Out – Rock, Paper, Scissors?


Some features are so core to your product or brand promise that you can’t operate an authentic business without them – they are your product. They slide easily into the IN column, no questions or doubts.

But as you’re building those features, you start to get little niggly paranoid thoughts that say “Well, what about this? If it were me, I’d be so annoyed if I couldn’t do that!”

It’s important not to be self referential at times like this. That’s not to say that you shouldn’t do a little “designing for yourself” – which can be a smart direction if you happen to be a relevant part of the target demographic coupled with being a subject matter expert in the field.

However, most of the time you are just projecting an unrealistic notion of how people will react to a feature they aren’t even yet aware of.

Sure, there may be a person out there, somewhere, who would love to have the extra little micro-feature you just imagined. But the truth is often that they’ll get by without it. People have a way of working around small annoyances, and developing their own way of making things happen.

Don’t let edge cases interrupt your progress.

The Evils of the Edge Case

An edge case is a problem or situation that occurs infrequently, under very specific circumstances, or something that is relevant only to a very small proportion of your user base. As hard as it is, these “edge cases” must be tucked away in a box for a rainy nerdy day.

An extreme example would look something like this:

Exaggerated Unbounce Edge Case Example

Scenario
A new customer – a traditional marketing company that creates bus stop ads – decides to move into the online marketing space. They have ads all over the city and want to get these converted into landing pages for a PPC campaign without having to employ a web team.

Feature that gets built as a result
An iPhone App that allows you to take a photo of the ad at the bus stop. You create a hot spot for a Call To Action link by dragging guides on the iPhone screen. It uses a visual recognition algorithm to separate text from background to create search engine friendly copy. Finally, it overlays a color mask on the CTA to generate multiple variants for an A/B split test measuring which button color converts better.

Brilliant? Perhaps.
Necessary? Of course not.

Assuming a reasonable facsimile of business sense – for something so absurd to pass the MVP screening process, there would also have to be huge fluffy flakes of snow raining down in hell.

The very nature of many edge cases means that their peculiarity can inspire wonderfully creative solutions, but they will result in a disproportionately small return on the time investment.

The problem is that for every happy edge case customer, you’ve missed the opportunity to create 100 happy regular customers.


When Edge Cases Stop Being Edge Cases

After 10 years at the top of the SaaS food chain, 37 Signals came to an interesting conclusion. When your customer base reaches a certain critical mass, what were once considered edge cases (occurrences that effect roughly 1% of your customers), now effect such a substantial number of people that it’s no longer ok to dump them in the acceptable minority bin.


As Unbounce approaches it’s initial launch date, the desire to solve all the known problems of our soon-to-be customer base, is huge. The product held in our vision is exciting, really exciting, and we can’t wait to get it out in front of people.

But at the end of the day, the goal of any product or service is to provide pain relief for an existing business problem.

Don’t leave your customers in pain while you try to fumble over the perfections of your product. Start with relief to the major organs then listen closely as they help guide you to where the next most painful spot is.

Oli Gardner

About The Author

Photo of Oli Gardner

Oli is Co-Founder & Creative Director at Unbounce. He's seen more landing pages than anyone on the planet. You should follow Oli on Twitter
» More blog posts by

Comments

  1. Carl Schmidt says:

    MVP is *hard*. You really have to get in the mindset that it’s ok for your peers and customers to see something that’s not everything you imagined it could be. The reward is that your product will ultimately be better than you could imagine.

  2. Great article Oli. We are dealing with these same issues as we near our launch. MVP is such a discipline when you have a vision of what the product should do.

  3. We are told to pursue niche markets and products. In the MVP world it seems that you are looking to serve customers who are looking for niche suppliers. We narrow down who we market to, your customers will be looking for enough service with little or no clutter. The problem is one mans clutter is another mans required feature. In a bell-curve the mean is only one person wide. To have a larger market you must accommodate the requirements of the persons to the right and possible to the left of center bubble. Those persons require a feature someone else considers noise. That one person wide mean will need to be hundreds or thousands of buyers deep. Clones of the one targeted mean, or noise must be allowed at some level.

  4. Duncan Ray says:

    Interesting, thanks! It’s worth looking at KANO needs analysis is useful for these kinds of considerations

  5. [...] Building a Minimum Viable Product – No Time For Edge Cases | Unbounce [...]

  6. [...] Building a Minimum Viable Product – No Time For Edge Cases | Unbounce [...]

  7. [...] Building a Minimum Viable Product – No Time For Edge Cases | Unbounce [...]

  8. [...] Building a Minimum Viable Product – No Time For Edge Cases | Unbounce [...]