Choosing Features for your Minimum Viable Product (MVP): Pooch-A-Tronix!
Dog owners need your Pooch-A-Tronix automatic dog feeder — they just don’t know it yet because you haven’t build and sold it. After canvassing friends, posting on Reddit, and talking to pet food store staffers, you’ve amassed a long list of possible features. A very long list.
A dauntingly long list of cool features you couldn’t possibly implement all at once.
You’re ready to create a minimum viable product: the simplest version of the Pooch-A-Tronix that will yield customer feedback. You want the MVP to get customers excited. Given the feedback, you’ll figure out what to build next.
But how do you figure out which feature from your list to implement first?
This one decision to choose a feature for your MVP seems hard to make. That one decision is actually composed of a number of smaller, easier decisions. It’d be a relief to make a bunch of small decisions.
Make One Small Decision at a Time
The Pooch-A-Tronix is an example of a grand dream that many entrepreneurs and builders have: a market-busting product ten times better than whatever is on the market now. To fulfill that dream you don’t need an elaborate plan. User feedback will guide you, but you should make a good first step.
For the Pooch-A-Tronix, the task to choose a feature for the Minimum Viable Product can be broken down into a bunch of small questions about each feature.
- Would the feature differentiate me from competitors?
- How closely does the feature align with my goals for my business?
- How frequently would the consumer use the feature?
Each individual question may be obvious in retrospect — that’s good! Simple questions often have simple answers. But we may be capturing dozens and possibly hundreds of simple answers.
Qualify Features using a Matrix
To capture the list of features and the answers to questions about each feature you can use a matrix — a grid. Let’s see how that matrix could make it easy for you to decide what feature(s) go into your MVP.
Your can draw the matrix on paper or create a spreadsheet.
Categories by which to Qualify a Feature
For the Pooch-A-Tronix we would consider each feature one at a time. For each feature, we answer one simple question at a time. To determine whether a feature is worth implementing, we need to consider whether the feature makes business sense, and whether it’s possible to implement the feature.
Here are the categories we’ll consider for the Pooch-A-Tronix:
- Differentiation from competition
- Feasibility of implementing the feature
- Concreteness of feature definition
- Legal liability
- Strategic value of the feature to your business
- Frequency of use of the feature
Differentiation from Competition
Ask yourself whether the feature is truly new. You may think you need to include a food bowl in the Pooch-A-Tronix, but the food bowl is not a differentiator. Every dog feeder includes a food bowl.
When we asked the question “Is this feature a differentiator?” possible answers are
- One of a kind: it doesn’t exist in the market [and the idea excites prospective customers]
- Unusual: the feature exists in few products in the market, but it’s not unique
- Common: many products have this feature
- Unwanted: the feature might seem cool, but no one actually wants it (This comes up more often than we often expect)
- Uncertain: honestly, you just don’t know the answer here. Acknowledge that and move on.
For example, if the Pooch-A-Tronix beeps when food is ready, you would acknowledge this is an unusual feature, but there’s a similar feature already available in the Acme Dog Fooderator and the Kaynyne Nine-Way Food Dispenser already in the market.
The screenshot below shows different possible answers to the Differentiator? question as a dropdown list. Okay, so I made the matrix kinda fancy here.
Green Means Go! If a feature tops out its category — it’s the best, the most, the grandest — then in addition to a short description of that quality such as “One of a Kind,” we also assign a color for sighted users that indicates the relative desirability of that state. For differentiation from competition, having a one-of-a-kind feature is the best state of all.
Each feature category has its own possible answers, but we use a consistent color scheme throughout:
- Green Means Go!
- Yellowish green indicates a feature that could be competitive, but isn’t unique.
- Orange indicates a problem. For example, if a feature is common, you should consider not implementing it in your MVP. You might only include the feature if it’s required to support other, more interesting features.
- Pink is a show stopper: no one wants the feature; it’s too hard to implement; or maybe it would be too expensive.
- Black indicates uncertainty and/or the unknown.
Feasibility of implementing the feature
Can you develop the MVP feature on your own? Or do you need a team? Or would implementing the feature, no matter how cool, be difficult even for a team?
Concreteness of Feature Definition
Is the feature concretely defined? “Beep when dog food is delivered to the bowl.” Choosing the right beep could take time, but the feature is clear.
How about a feature that’s too vague? For example, “Compare the food (the input) going into my dog’s front end, and compare this to the quality of output coming out the other end.” That could be a boon to dogmothers and dogfathers everywhere, but how exactly would this be implemented?
Sometimes working through your list affords an opportunity to add another item to the list that is more clearly defined. Keep both the original feature description and the improved feature description in your matrix. Don’t prematurely remove a feature from your list. Your matrix will serve as a historical document, and something you could even present to customers, team members, and your future (wiser) self. Document your journey.
If you sell anything for any amount of money, no matter small that amount of money, then you’ll need to consider the legal liability.
Your feature will fail at some point. What are the consequences?
Although you can talk to a lawyer about the cost of liability insurance, you should first consider whether legal liability could threaten your product and your business. High liability isn’t necessarily a show stopper. Could Pooch-A-Tronix make a dog sick?
Is this feature critical to the long-term plan for your business? It would be cool to diagnose a dog’s feeding problems. However, if you want Pooch-A-Tronix to be The Fun Feeder rather than The Diagnostic Feeder, diagnosing feeding problems isn’t key to your strategy.
Your product will not be everything to everybody. You can not be the best in every possible feature in your list. If you identify too many features as Critical to your strategy, your strategy may not be clearly defined.
Frequency of Use
If a feature won’t be used at least once a day, consider not including that feature at all.
If a consumer might only “need” a feature once a week, assume that feature may never actually be used. In the 80s and 90s, few people knew how to set the clock on their VCRs. The interfaces were terrible, sure, but setting the clock had nothing to do with playing movies.
Daily use means daily testing for your MVP. That’s another reason to prioritize features that would see daily usage: you want feedback soon.
All Pooch-A-Tronix Features Qualified
In the screenshot below, fourteen features have been qualified in our six categories. There’s a Notes column to provide a little extra info.
Examine each row. What sticks out?
Look for “Green Means Go!” All the Way Across
Once you’ve filled out your spreadsheet, the Big Decision becomes simple: check for Green Means Go indicators all the way across.
If you’ve identified a feature that is one of kind, if you can develop it yourself, if the feature is defined very concretely, legal liability is low, the feature fits your strategic plans, and the consumer would use the feature daily, then you’ve just identified a feature for your MVP.
Dog Food Catapult.
DOG. FOOD. CATAPULT!!!!
For the Pooch-A-Tronix feature list there is a clear winner: Dog Food Catapult! Green all the way across.
For the sake of example I’ve set the legal liability for the Dog Food Catapult as None. That’s inaccurate. It’s easy enough to imagine dog food landing on your dinner table, in Fluffy’s ear, or perhaps in a customer’s water glass. Okay, maybe there should be a column to indicate how much fun the product is both when it works AND when it doesn’t work.
Eighty-four Questions and Answers.
One Clear Decision.
Breaking down the question “What feature should I implement first” lead to no fewer than 84 questions and answers.
14 features * 6 questions per feature = 84 answers
Each one of those 84 answers takes a little thought.
That we made a decision by answering dozens of questions helps explain why choosing a first feature for your MVP is a hard decision to make if you think of it as just one question. And also consider whether it would have been possible to have eighty-four answers in your head at once.
Pooch-A-Tronix?!? Um … what about Food-A-Pult?
Identifying the first feature for your MVP could also lead to rebranding your product.
“Pooch-A-Tronix” would have been a cool product in 1982 when E.T: The Extra-Terrestrial was in theaters and just about anything electronic was cool.
But it’s 2021, and “-Tronix” doesn’t tell anyone anything interesting about your product. Your product launches food in the air. Your product is fun for you and for your dog.
If your product is a fun food catapult, maybe … Food-A-Pult? That name tells you something right there.
And now you have a method to determine whether that name works. Create a matrix! Look for Green Means Go all the way across!
- Gather a bunch of ideas for names.
- Brainstorm a list of categories by which you could qualify the appropriateness of each name: uniqueness, strategic fit, etc.
- Rate each name for each category.
- Look for green all the way across!
Monday.com: Spreadsheets, Interactions, & Integrations
For project management I use Monday.com. The selection matrix for the Pooch-A-Tronix is adapted from a Monday “board” I created in 2020 to identify what my MVP should be.
Monday.com is simple to learn, it supports integration with numerous other tools and services, and a few simple interactions are used throughout the product. Their customer service is fantastic.
I’m just a Monday.com user, and I certainly don’t get paid to mention them. They’re popular already, and for good reason.
But to create your Food-A-Pult, all you really need is a method and a piece of paper or a spreadsheet.