Over the last 25 years, I’ve worked on the launches of 23 different products, in many markets from medical to clothing. I’ve watched with pride as our launches have hit huge numbers in sales, and worked with some of Silicon Valley’s top names to help them design killer MVPs. There have also been failures along the way and of course it’s our failures that teach us the most.


The main question most entrepreneurs have is ‘what makes the difference’? What predicts the success or failure of a product, and how can you design an MVP that ensures you get that all-important early traction?


Having reviewed product the launches and MVP I’ve been involved with, I’ve distilled the process that the most successful followed. In this post I’ll be sharing it so that you can design the MVP for your next product.


Bringing products to market is probably one of the most challenging but also fun jobs to have as you need to touch and work with almost every department in a company from R&D to engineering, sales, legal, support and of course finance. There is no better training if you want to be a CEO.

I have my own “way” of defining product specifications which I’d like to share by way of a process to define the minimal viable product (MVP).

This may raise more questions than answers for some, so for a free MVP review contact me at annie@magicmonkey.eu or download the MVP Checklist@  www.downloads@magicmonkey.eu

So I am going to talk about a number of topics relating to defining an MVP.

  • What is the Minimal Viable Product? (MVP)
  • How do you go about defining the MVP?
  • How can you come up with ideas for the MVP?
  • How can you validate that you have identified the MVP?
  • When do you freeze the MVP spec?
  • What are the things that can go wrong in defining the MVP?
  • How long will it take to go through the MVP Process?

Maybe you have been there? After months maybe years of product development the day comes when the product is launched to a lukewarm reception. You offer launch discounts, bundle other products with your new offering, yet the customer is not enthusiastic to buy.

You don’t get it. You interviewed them months ago sharing the idea for the product perhaps even the spec and the interviewees said it looked good and there was a niche in the marketplace. Yet now when it’s here they don’t want to give you the most precious of all objects – the purchase order.

So how did it all go wrong?

Maybe you have developed what you thought the customer asked for but haven’t created a minimal viable product?

What is the Minimal Viable Product?

The minimal viable product or MVP is a product that delivers the minimal feature set and benefits to solve a specific problem.

The most important word in the above sentence is “minimal” as if you have delivered much more than the MVP with features and benefits the customer does not want or need, it’s likely to have taken longer and been more expensive to develop than it needed to be at its launch.

So the trick is to identify the MVP and build it and ONLY it first. Then if you want to enhance the product as a result of future customer input add those features later.

How do you go about defining the MVP?

There are 20 steps to follow to define the MVP. You could of course just go and ask potential customers what they want in an informal chat but this typically doesn’t work for a couple of reasons:

  • They won’t know what they want, and typically ask you what you have!
  • You won’t be able to compare and rank various customer responses, so it’s best to have a structured process as outlined below:


STEP 1. Draft the Datasheet

Draft a product datasheet (let’s call the product P1). The datasheet describes the product as it if actually existed to include:

  1. Product description
  2. Features
  3. Benefits
  4. Appropriate technical information
  5. Three reasons why the prospect should buy P1 starting with

“You should buy P1 If…………………………”

It actually helps hugely to do this exercise as it makes the product “real”, so it’s helpful to have a product name. If you can’t decide on a product name give it a codename. You can change that later before launch. If you are struggling to name your product download

“How to Name a Product the Ultimate Guide” from  www.magicmonkey.eu/downloads

STEP 2. Competitive Analysis

Put the features of the proposed product (P1) into a spreadsheet

Complete a competitive analysis which shows the features of your product compared with the features of all of its competitors. This might seem like a lot of work but it will be worth it as in STEP 7 below if the customer starts to compare P1 to its competition you will be in a knowledgeable position to deal with any queries or objections.

Most of the data you will need you will find on competitor websites. This exercise should also show you your USPs (or lack of them!). If you are not sure how to do a competitive analysis download:

“Ten steps to a really useful competitive analysis” (Word doc) from www.magicmonkey.eu/downloads.

STEP 3. Build the Product PowerPoint.

Make a short PowerPoint of P1 that you can walk through with prospects face to face to explain your vision for the product and its features and benefits. Think about workflow. Show how it would be used in the PowerPoint. If it has a user interface mock it up in PowerPoint putting the brand name of the suspect into it if appropriate so you begin to get the customer to think of the product as “his”.

STEP 4. Pricing First Pass

Have a general idea of the price you wish to charge, or a price range. This means knowing what it will cost to produce – its BOM (Bill of materials).

STEP 5. Prepare the Interview Template

Create a table/spreadsheet with the most important features of P1 at the top as shown below. So in the template, you think “FRED” is the most important feature. You may also think FRED is a USP. Add columns that reflect potential interviewee feedback. In this way, we will be able to compare feedback from various interviewees.

FEATURE Features Description Don’t need it 0 Nice to Have 1 Beneficial 2 Essential 3 WOW 4  I Don’t believe it!

STEP 6. Get a NDA Signed

Once you have done this you are ready for your first suspect interview, so you will probably want the suspect to sign a non-disclosure agreement.

Note I say “suspect” here not customer. I like to put leads into 4 categories we work with in a sales scenario:

  • Suspects: these are people we think will buy but we need to validate they have the capability, money, authority and so forth
  • Prospects: these are people we have validated to have the need and capability to buy
  • Customers: these are people that have already bought our products and services
  • Evangelists: These are people who have probably bought our products and services and encourage others to also buy. These are a valuable corporate asset. For a free copy of my book “Dream Ticket” which tells you how to grow valuable corporate assets contact annie@magicmonkey.eu

At this stage we have not validated that the interviewee is a prospect for P1 (he might be a customer of an existing product but not for this one yet).

STEP 7. Meet with the Suspect and show the PowerPoint

Sit with the suspect and give him an overview of P1 with the PowerPoint. If possible do it 1-1 with the PPT on a laptop so its “intimate”.

STEP 8. Give the Prospect the Datasheet

Give the suspect the datasheet and let him read it for a few minutes so he sees the full proposition and can ask questions.

STEP 9. Explain the Ranking Categories

Then, sharing the interview template with the suspect walk him through each of the ranking categories.
NEVER give the suspect the form below to complete on his own. The form is our interview template. We will be asking the suspect about his responses to get more information from him, asking “why?, can you explain more” probing whenever possible.

Then taking the template explain the categories.


  • Don’t need it
    • “This feature is of no interest to me”
  • Nice to have
    • “I might use this feature, but it’s not that important to me”. “Nice to have” is never incentive to buy.
  • Beneficial
    • “Yes I can see that having this would be useful” so he is saying it’s not essential.
  • Essential
    • “Don’t bother making the product unless it has this feature in it” so this feature should be towards the top of the feature list.
  • WOW
    • “Can you really build this?” it’s a positive reaction to a feature he probably thought he could never have, or have in the timeframe or budget.
  • I Don’t Believe It
    • “I don’t believe you can build or deliver this feature in this product”

This happens when the suspect does not believe you have the ability to deliver this feature OR he does not believe it’s possible for anyone to do it.

It’s a real problem when this happens as there is no point getting into a “We can” “no you can’t” debate with the suspect. So to get him to buy you will have to convince him somehow that this feature is deliverable. Even then he may still not believe you. Products that have ranking in this column tend to be disruptive and these come with a host of problems to solve in order to sell them.

STEP 10. Categorise the Features

Go through each of the features with the suspect and ask him to categorise them.

STEP 11. Fill in the suspect’s responses

No Features Description for ROBO Don’t need it
Nice to Have






  I Don’t believe it!

STEP 12. Sort the Responses
Next sort the responses on-the-fly circling the 3’S AND 4’S and then put it back to the suspect that: “ If we made this product with the features you have scored as 3s and 4s how much would you pay for it? (You may get a nice surprise!)

STEP 13. Validate 1 Ask for the Order 1!
Then ask the suspect “If we made this product with the features you have scored as 3s and 4s would you buy it in X months at Y price?

STEP 14. Investigate Objections
If he says” No” determine what the problem is price, time, features etc.

STEP 15. Add proposed Missing Features to the MVP
Ask the suspect, “What features are missing from this template you consider to be essential before you would make a purchase decision, or are a WOW factor for you?” add those features to the template under “Missing Features”

STEP 16. Validate 2 Ask for the Order 1!
Then ask the suspect “If we added these missing features would you buy the product at this price?”

STEP 17. Investigate Objections 2
He may disagree with the price or timeframe which is your opportunity to go back and quiz him about what he really wants by what date.

STEP 18. Identify Disruptive Features
If he has a “YES” in the “I don’t believe it box” and it’s at the top of the feature list (which it probably will be) you have a problem. As he is saying that he does not think you can deliver/build the feature which is likely one of your USPs.

STEP 19. Move the “I don’t believe it” feature”
Lastly, you need to have a dialogue with the suspect about how to get over the “I don’t believe it hurdle”. Typically this means a try-and-buy scenario, a trial or freebie.
The only thing you must not do is ignore “I don’t believe it” as shifting the “Position” in the mind of the suspect or prospect “I don’t believe it” is very difficult and will add a great deal of time to the sales cycle. I call this the VOID which I first wrote about in 2011 in a paper
“The VOID and the Chasm: Planning and Developing Products that use Disruptive Technology” www.magicmonkey/downloads
You don’t want to fall into it!

People who don’t believe won’t buy until the end of the product adoption curve (late majority, laggards). This can and has taken years with a couple of products I have brought to market.

STEP 20 Collate Missing Features

As each suspect interview is completed add the Missing Features to the bottom of the subsequent interview template. In this way you can see if there is a majority for suggested additions the suspects categorise as Beneficial or Essential or it’s just a one-off preference for a particular customer/organisation. Be wary of “the big customer scenario” so that you don’t build a product that only has one customer.

An Example

The prospect CAROLINE is looking at the features of the product code-named ROBO which has 10 features so a potential score of 30 to 40 for the MVP (we only care about Essential and WOW at this time). She can only chose one category for each feature.

No Features Description for ROBO Don’t need it
Nice to Have






  I Don’t believe it!
  MVP TOTAL = 21


Analysis of Caroline’s response

  • She doesn’t need FRED or BRUCE
  • She can see the benefit of having JOE but’s not essential
  • CHARLES, HARRY AND PAUL are essential
  • DAVE, MARTIN AND NIGEL impress her greatly
  • She does not believe SIMON is possible to have for a reason you will need to discover. Then see if you can persuade her that you can deliver it and move that feature into the WOW column.


Adding up the score for this feature set its 21, which would indicate a marginal spec that had features this Caroline did not care about.
Next you need to repeat the exercise several times with different suspects, asking about the Missing List IF the product team thinks these proposed features are viable.

How can you come up with ideas for the MVP?
There are several ways to come up with ideas for the MVP.

  1. In-house brainstorm with the product team
  2. Creating an Innovation Team within the company
  3. Lots of dialogue with prospects and customers “if you could have ANY features in P1 what would they be?”
  4. Tracking the competition
  5. Talking to industry experts and luminaries
  6. Having an advisory board
  7. Tracking academic research
  8. Stealing ideas from other industries and adapting them to suit yours


How can you validate that you have identified the MVP?
In truth you can’t validate that you have nailed the MVP until you have healthy sales. However using the 20 step process above with a suitable suspect/prospect set – say a minimum of 10 suspects depending on the nature of the product, you should get a really good feel that you are building something that prospects will buy. That’s of course as long as you listen to the voice of the suspect, which sadly does not always happen when development teams think they know better.


What most always does happen is the suspect tells you what they don’t want and what is not important to them. If you take heed of this information quickly there’s an opportunity to rethink and re-present a product solution that ticks the boxes of an MVP.


When do you freeze the MVP?


Once you have gone through the 20 Step MVP process and have summarised your findings and they indicate future market pull it’s time to determine whether or not you can build it for the indicative price. This will mean preparing a requirements analysis so that the design/engineering team can respond with a functional specification, timeline and cost. Then of course you need to ensure the budget is available.


Beware of feature cutting to squeeze a specification into a budget. Only build the MVP.


What are the things that can go wrong?


If you follow the 20 Steps to a MVP rigorously you can’t go far wrong. Of course the prospect could mislead you and say they will buy when they won’t, or don’t have the authority to purchase, but the way to mitigate this risk is to:


  1. Make sure the set of suspects is large enough so a couple of bad eggs don’t skew the data.
  2. The feature set the suspects want may be radically different from your proposal, meaning that the cost to develop what the suspect wants is much greater than the available budget. The solution to this crisis is:
    1. Make sure you are interviewing the right people
    2. Get a bigger budget
    3. Don’t build the product!


How long will it take to go through the MVP Process?


Of course, this depends on the nature of the product. Specifying an airplane will take longer than a cancer scanner, and it depends on how many different user types you have that may use the product in different ways and so have different workflows. Assuming you have 3 user types/workflows and wish to interview 10 prospects in each group that could be done in a couple of months. For some that may seem like a long time but the value in getting the MVP right first time is immeasurable maybe saving months if not years of development time.


The process of suspect interviewing also yields other benefits.


  1. You may be able to get a purchase order against delivery of a product to a specification. If the product is something the suspect (now a prospect) really wants you may be able to set up a scenario where you say “If we deliver this product to this specification by this date can I have a Purchase Order now?” 

    The prospect is taking the risk that a competitive product will arrive in the market in the time he is waiting for yours to launch. Or that his need will go away.


    From your perspective you have nothing to lose. If you fail to deliver the prospect is not committed to buy. If you deliver to spec the customer should honour the PO with the additional bonus that you have some evidence of pull in the marketplace which is a great comfort.


    A sales backlog is a corporate asset. However don’t ignore him in the development period, keep in touch, make sure he knows the product is coming and he still has a need for it. For a free copy of my book “Dream Ticket” which tells you how to grow valuable corporate assets contact me at annie@magicmonkey.eu and I will send you a free copy.


    I once had to help a US VC client who had invested $1million in a company that had a large PO to develop a novel battery specification for a new laptop. Unfortunately the CEO of the battery company did not communicate with the customer during the battery development period. Arriving months later with a working prototype he discovered the form factor for the laptop had changed and the battery did not fit!

  2. Identifying a group of prospects may also give you the opportunity to bring them with you on the product development journey. They may be your alpha testing sites. You could loan early versions of your product and observe the prospect using your product in order to validate the workflow and make sure it fits into the environment where the prospect will use it. This may give you some early case studies.A group of prospects can be turned into a focus group that meets from time to time getting updates on product development. In return they may suggest ideas for features that may be added to subsequent product versions. You may give early bird discounts to prospects if they purchase on launch day which in turn enables you to generate press releases on a successful launch – with sales.
    If you would like some help planning for a product launch download
    “10 Things you must do before launching a new product” from www.magicmonkey.eu/downloads
  3. You can work with prospects to define the user interface.
    Using a PowerPoint presentation you can give the prospect a “simulation” of the experience of using the product. This will enable you to get early feedback on usability, flow and number of clicks to solution before you build the requirement for the UI software.


In summary getting the MVP right before a penny of development money is spent is the goal. Imagine how it would be if you had:


1.A validated specification with identifiable USPs
2.Workflow validated so you can see how the product would be used in its target environment
3.A validated user interface without spending any effort on software development
4.A group of early adopters ready to buy
5.Some case studies
6.A sales backlog
7.A focus group
8.A few pre-orders with PO’s pending delivery to a spec
9.Partners ready to do trials, alpha and beta testing.
10. A validated sales price with which to do sales forecasts


Not a bad shopping list for a new product or service before development has begun!!


For a free MVP checklist go to www.downloads@magicmonkey.eu