Making Sure You Are Ready to Begin Building Your MVP –
Study this article By Dr. Tony Karrer, use the links and you will be far ahead in building an MVP that actually does what you are expecting. The great thing about Tony’s blogs, as we have featured before, is that they are filled with links to all the other “terms” and “technical details” that you may need more information on. Dr. Karrer writes in his blog on April 2, 2013: In the presentation next week at Coloft (Details/Registration) I’m going to be looking at aspects like:
- Things to consider before building your MVP
- Features often overlooked when documenting an MVP for developers
- Understanding important metrics you want to measure
- Risks and challenges in developing an MVP.
It should be a fun evening with lots of interesting conversation. This post will provide links to participants as well as to readers.
What’s Going to Go Wrong
A lot of founders don’t really understand Lean Startup principles. They look at the following high level definition of Lean: and they interpret that as write up an executive summary with your ideas and hand it to developers to build. What’s going to go wrong? Well I often get the unfortunate call from startup founders where all kinds of things have gone wrong:
- Built the Wrong Product
- Poor Product Quality – Code is really bad, full of bugs, missing critical features
- Doesn’t Scale Past a Couple Users
- Not Protected – Access to Code, Rights on Code
Why are you building your MVP in the first place? See Investors, MVPs and Evidence of Traction. Have you conducted Problem, Solution and Feature Interviews with customers? Have you Documented Your MVP for Your Developers? Have you looked through Things You May Have Forgot in Your MVP and provided answers to these? Do you have a Technical Advisors: Every Web/Mobile Startup Must Have One? Don’t be fooled by a Common Misunderstanding in Agile Software Development.
While You Are Building Your MVP
Look for the following Symptoms of a Weak Development Team. Correct your course quickly. And when you have Poor Software Developers – Pull the Plug Early. Learn how to test and possibly use testing and load tools. See: http://www.softwareqatest.com/qatweb1.html
- How to Hunt Programmers for Your Startup – A Field Guide
- Finding a Technical Cofounder for Your Startup
- Choosing a Programming Language and Framework for Your Startup
- Startup Founder Developer Gap
- Startup Software Developers
- Technology Roles in Startups
- Startup CTO or Developer
- Startup Metrics
- Building Your MVP as a Non-Technical Startup Founder
Well I’m currently working with startup founders on their systems in JRuby, Django, PHP and Java. Several of these are B2B applications with relatively smaller audiences. I’m feeling okay about our choices of frameworks that are slower and will cost more in terms of hosting and managing growth. The availability of talent was an important factor. However, startup founders who are building applications that have:
- Large Audiences – consumer facing
- Complex Processing – examples: Matching, Social Network Analysis, Compatibility Scoring, etc.
- Database Intensive
need to consider going with a higher performance solution. Most startups do not get a chance to move from one framework to another. It takes a lot of time and effort and the result is that you get to go through a whole new set of bugs only to get back to where you started but with a faster, more scalable application. Think about twitter – but they had lots of money. Often we justify building an MVP in whatever framework is fastest to build or where we have resources that know that framework. You may get into the market, but just know that you are going to pay the price when you start to get traction.
“I have a few investors interested but they want to see a product.”
In Building Your MVP as a Non-Technical Startup Founder, I mentioned that before you build your Minimum Viable Product (MVP), you need to be really clear on your purpose. In most cases, when you are building your MVP, you are trying to prove out certain startup metricssuch as:
- Cost of Customer Acquisition
- Conversion Rates / Pricing
- Viral Coefficient
in order to get those numbers in front of investors so that you have evidence of traction and can show that you can begin to build out more of the real product and begin to scale the business. It is almost never the case that you are building an MVP to “show” to an investor the product itself. Yes, the investor may literally have said to you:
“That is something I’d seriously think about investing in when you have your product built.”
But the reality is that they don’t mean that. Two aspects to this:
- You can let an investor see your product via a mockup or clickable prototype.
- If you do build the MVP and show it to them, they will ask you about your metrics. They really want metrics, not a product.
A Mockup is Enough to Show the Product Most investors can look at a mockup or clickable prototype and have a pretty good sense of the product. They may wonder if it can be built technically, but I (or other CTOs) can answer that question without building any code. Cases Where a Mockup is Not EnoughThere are a few cases where mockups or clickable prototypes may not be enough:
- Usability, interaction design, etc. For example, the iPod won not because of better features and functions. It won because of interface, ease of use. To get an investor excited about another MP3 player at the time, they would have needed to play with the interface. That said, you likely could have still come up with some cheaper way than building the iPod.
- Results of Algorithms. Often you can’t tell if something like a search engine or matching algorithm is really going to have better results until you use it. You may have to build something out to show it working for someone to evaluate whether it really works better than what else exists.
I would guess that this represents less than 5% of startups. Investors Really Want Evidence of TractionSo you built your MVP; you bring it to the investor; you demo it; and I will guarantee they will ask you:
“So how many users do you have? How much is it costing you to get users? How much are you making from your users?”
And other similar questions. Yes, they are happy that you have your product built and that does make it much more investable. But now that you have a product, you should be able to show that people want to and/or are willing to pay to use it. Bad News? Not ReallyAt first you may be thinking that this is all bad news. Wow, now Tony is telling me that not only do I need to build my MVP, but I need to actually show evidence of traction. That’s an even tougher hurdle. Yes, that’s correct. Sorry, but that’s the reality. However, the good news is that for most investors, you can certainly change the question and get a lot more information without ever building the MVP. The real question you should be asking is “When I’ve built this product and show you the following metrics, would you invest?” The “this product” will be a well formulated mock-up or clickable prototype. If you don’t have that, then you will naturally let the investor off the hook by saying, show me the product. I actually think you can push the conversation pretty far with most investors and a few good mockups. No, you won’t know if you will really get a check – most investors are hard to actually get a check from – but you will have a pretty good indication. Bottom line is that before you go and build your MVP:
- Define your MVP really well on paper and Document Your MVP for a Developer
- Do Your Homework Before You Develop Anything
- Make Sure You look at Questions Developers May Have Forgot to Ask a Startup Founder
- Figure out the Evidence of Traction you really need
- Get a Technical Advisors: Every Web/Mobile Startup Must Have One
- Test all of this with Investors
How Much Traction is Enough for Investors?But never forget that traction is necessary, but may not be sufficient, to lower the risk perception of investors, and assure an investment. The quality of the team, and overall financial health are equally important, as well as how your offering compares to competitors. Now that You’ve Got MVP, It’s Time to Think About MVCYou can’t raise money on achieving an MVP. Investors demand more than that. As Steve Blank likes to say:
A Startup Is a Temporary Organization Designed to Search for A Repeatable and Scalable Business Model
The unfortunate reality is – an MVP is not the above! Yet most of the newly minted entrepreneurs I’ve met think their job is nearly done when they’ve found MVP – they think they can go build a pitch off their early MVP and raise money! A startup does require MVP but it is much more than just MVP. The problem is that MVP means early adoption of product and its features, maybe even some who will pay. But it doesn’t tell you how many people will do it in the long term and whether this can support the company (the people and operations within) that is behind it. What The Heck Does “Traction” Really Mean To A VC? Great analysis from John Greathouse: Do You Speak VC? A Beginner’s Guide to Investors How to Speak the Language of Venture Capital
Wednesday, February 13, 2013
Purpose of an MVP and Defining the Right MVP
I’ve really not talked as much about this in my blog even though its hugely important. I generally hear the following as reasons for founders building their MVP:
- I need it to get investors interested (NOT VALID)
- I need it to see how early customers use it to get feedback (RARELY VALID)
- I need it to test/prove aspects of the product such as (VALID)
- Cost of Customer Acquisition
- Conversion Rates / Pricing
- Viral Coefficient
My claim is that the first bullet, although extremely common, is a misguided reason to build the MVP. Investors my tell you that, but what they can look at your product on paper and tell what it does and they will understand if it can be built. Once you build it, they will now ask you about the key metrics that they need proven in order to see if you really are a good investment. The second bullet, getting feedback from customers is most often not valid either. Again, putting something down on paper (wireframes, graphic comps) and getting feedback from potential users can tell you most of what you would learn from a working MVP. There are a few cases where you somewhat need to see the system operating to have a sense of the value. Examples might be a recommendation engine, search engine, matching engine or something with a complex interface. Even with these, you will have paper-tested your MVP, but the reality is that customers will not be able to assess the value to them until they actually use it. The real reason to build an MVP is to do early tests of key Startup Metricsfor the business. To prove/disprove a hypothesis. One of the questions in the session was “How do you know what to put in your MVP?” Once you have the metrics defined, it focuses your effort.
Ways to Make Your MVP More Minimum
We spent quite a bit of time talking about a complexity scale and the kinds of resources you can viably use at different levels of complexity. Simple MVPs get built with less the 3 Programmer Months worth of effort (that’s 3 months a single programmer working full-time). More complex MVPs are going to be 12+ Programmer months. There are some MVPs that are unavoidably complex. eHarmony for me fell into that camp. We needed the matching algorithm. Many MVPs can be made to be very simple.
- Paper Prototype or a Smoke and Mirrors Prototype – you can just build something on paper or even just one that you can click through and get a lot of the feedback you need.
- Fake Site – you can have what looks to be a real site, even take “orders” but not actually have anything able to run it.
- Leverage Existing Platforms or Third Party Products – you want to test your social network, grab Drupal and whip something together, or even just use a hosted service.
- WordPress – we spent quite a bit of time talking about how you could do a lot with WordPress to provide simple forms of lots of functionality. WordPress is pretty easy to hack. And the back-end is something that a non-technical founder can manage. We end up using WordPress a lot as the marketing front-end of our web sites.
The bottom line is that you should first look to make your Minimum as Minimum as possible. Here are some important and relevant posts that relate to the topic of defining the right MVP. The “Questions” post is probably the most important.
- Don’t Subtract – Restart to Find the Minimum Viable Product
- 32 Questions Developers May Have Forgot to Ask a Startup Founder
- Startup Software Development – Do Your Homework Before You Develop Anything
- Document Your MVP for a Developer
What Do you Need to Get Your MVP Built
From the graphic above, you can see the kind of resource you might need to be able to build your MVP. If you are on the lower complexity end, the key is defining small chunks of work that can be done quickly by a developer. If you do not break it down into small pieces, its hard to make progress with part-time resources, freelancers, etc.
- Equity-Only CTO and Equity-Only Developers
- Technology Roles in Startups
- Want to Know the Difference Between a CTO and a VP Engineering? (someone asked specifically about this)
- Hiring a CTO for Your Startup
Founder / Developer Gap
I’ve spent a lot of time on the Startup Founder Developer Gap and knowing what you need in terms of a Startup CTO or Developer. In this talk, we spent most of our time on Technical Advisors: Every Web/Mobile Startup Must Have Oneand how they should be helping you:
- Specify the right things to be built.
- Third party products are used appropriately.
- Structure development contracts appropriately or directing the in-house team appropriately.
- Plan for past the initial MVP.
- Review the code being built.
Finding and Selecting a Technical Cofounder / Developers
- How to Hunt Programmers for Your Startup – A Field Guide
- Finding a Technical Cofounder for Your Startup
- Startups and a Common Misunderstanding in Agile Software Development
- Poor Software Developers – Pull the Plug Early
- Symptoms of a Weak Development Team
Choosing a Programming Language and Framework for Your Startup Finally, if you made it this far, then take a look at: Free Startup CTO Consulting Sessions. I’m always happy to try to help startups figure out how to go after creating their MVP.
Wednesday, January 2, 2013
I’m leading the marketing efforts for an early-stage startup. I recently completed an MBA, which I feel gave me a good basis in the fundamentals of building our brand from the ground up. However, I often wish that I had someone more experienced both in marketing and working in a startup environment that I could go to for advice. Do you have any suggestions for how to find a good mentor?
Great question and I believe that just like finding a Technical Advisorfor your startup is critical, finding a good mentor is critical.
Mentor vs. Advisor
Generally when I talk about startup mentors and advisors, I distinguish them in terms of:
- Mentor focus is You:The marketing person above is looking for someone to help them with their personal challenge. They want someone who understands their current role and can help them be successful there. They will also be concerned about how this leads to their next role. They may open doors to the next role.
- Advisor focus is Your Business:The startup founders need help with the business and I’m advising them to find an Advisor who will focus more on the business then on them as a person. They likely want someone who knows the industry and can open doors that can help the business.
These are not mutually exclusive and good mentors and advisors get into both.
Finding Potential Mentors
You are looking for someone who is
- A couple steps ahead of you in progression
- Have worked in similar roles in similar kinds of companies
- Local to you
For example, I’m looking for a marketing professional who’s worked at a couple early-stage startups ideally one of these is B2B selling to advertisers and is located in Los Angeles area. You should begin asking at networking events and people you know in the startup world. For me, I’d use LinkedIn. The challenge is that LinkedIn is not that great with finding people who were at early-stage startups. So the way I would do it is to find examples of specific startups that have grown up locally and who marketed similar to how you plan to market. Then you search for the people who were in marketing there early on. Likely they’ve progressed, so it may be a pretty natural fit. This can take some work, but its worth it. This also avoids a common problem that will sometimes happen when you go the networking route. People will suggest folks who are high profile. They regularly speaks at conferences, used to work at Google, and have other similarly impressive aspects to their background. That may sound good, but that doesn’t make them a good mentor. Have they worked in similar kinds of situations? Most of us are not Google. An early-stage startup is a different animal. From Why Every Startup Founder Needs a Mentor – And How to Find One
Don’t just fall in love with someone’s reputation, perceived celebrity or name. Identify someone who could be directly relevant to what you want to do, or who is pursuing a similar vision. And someone who is likely to have the time and the inclination to help you.
To me this is a complex problem, it’s a bit like dating/marriage. You are the person who wants to commit but you have no idea if the other party is going to be willing to commit to you. I don’t believe that you should go out saying to a potential mentor that you are looking for a mentor. While that’s open and honest, you know how it works out if you bring up commitment too early. And I’m not alone. In How to Find and Keep Your Ideal Mentor:
Once you’ve identified an ideal mentor, you need to create the relationship that houses the mentorship. Here’s a little secret about finding mentors — nobody has time to mentor you. And they probably lack interest initially. If you’ve ever sought out mentors, you’ve probably found that many of them were reluctant to make the commitment and ran when they heard the word “mentor.”
So, where do you start? For me, its coffee?
Can I buy you a coffee and talk to you about my challenges?
Or in the case of the LinkedIn outreach:
Hi John, I’m leading the marketing efforts for StartupRoar, an early-stage startup that sells to B2B advertisers. We are doing well, but have some interesting challenges. It looks like you were in a similar kind of role at XXX and you likely had some of the same challenges. Would you be open to getting together for coffee and talk about my challenges and how you addressed them before? Thanks, Tony
If you are working on something interesting and you ask nicely, there are a lot of people who are willing to have the conversation. And sometimes that grows into more conversations. If you’ve talked to a bunch of people and no one is willing to get coffee with you, then that’s a sign that you are working on something that’s not interesting or you’ve not done your homework. At the end of your first coffee, if things have gone well, I like to say
This was great. I really appreciate your time. I’ve got a lot of great ideas and things to do. Would you be open to getting together again at some point as I continue down this path?
That’s right, you begin to set up your second date right at the end of your first date.
Accelerator / Incubator Mentors
Most accelerators and incubators have a long list of potential mentors. I’m a mentor at Start Engine and Founders Institute LA. I want to warn you that just because you are going into one of these programs doesn’t mean you will find a mentor. It means you have easy access to people who could become a mentor. You still have to do the work of connecting with these people and turning them into an actual mentor. In Startup Mentoring, The Socratic Way, Fred Destin tells us:
The faster you can get past the pitch phase (“let me convince you that my idea is great”)to the constructive phase (“let me exchange with you to make this thing better”), the more benefit you will both derive from the interaction.
I’ve been in lots of conversations where the founder is just pitching me. This happens a lot at the incubators / accelerators. I’m thinking I’m in the meeting trying to help. All we get are pitches and most often there’s not opportunity for a startup to say, “I could use some help with X.” We all have challenges – lots of challenges. Be open about your challenges, especially when you are talking to a potential mentor. You still need to tell them about your challenges and start with a meeting/coffee. Let me stop here, because if you are in a program like this, then just go over to Ben Yoskovitz’s post How to Maximize the Value of Mentors in Accelerators.
Do Your Homework
Everybody’s equal because we have 24 hours. The question becomes, “What do you do with your 24 hours?”
He’s talking about making life choices, but anyone who’s worked in a startup knows that there are always a million things to do and we constantly make choices about where we will spend our time. The same is true of your advisors and mentors. So show respect for their time by doing your homework before you ask them questions. I wrote about this a long time ago in Questions Before You Ask. In that case, I was talking about the homework you should do before you reach out through LinkedIn for a question. There’s a ton of information out there, so if you have a particular situation, make sure you’ve:
- Searched for Answers
- Keep a List of What You’ve Found
- Read through What You Find
- Compose What You Find Into a Preliminary Answer
- Figure Out What the Real Question Is
- Ask the Real Question
If you are asking your mentor questions that shows you’ve not done your homework, you will likely lose access to that person’s time. They will push you to do your homework the first time. The second time, say goodbye.
- Venture Hacks – Advisors Series
- Finding (And Keeping) The Right Advisors For Your Business
- What is a Mentor Cycle
- Mentoring the mentors: Advice and inspiration for startup mentors
- How I Try to Mentor Startups (And Hopefully Add Value)
- Mentor Manifesto
- Find and engage great mentors
- Startup Mentoring Sessions – How to Get the Most Out of Them
Wednesday, December 19, 2012
Me: Do you have specs? Founder: Ummm … <embarrased look> … what do you mean? Me: Product definition, use cases, feature list, wireframes, comps, really whatever you have. Founder: Umm … <still embarrassed> … what format would you and the developer want that in?
I’m never trying to embarrass someone. I know how it feels. It’s the same as when I’ve created financial models and then have it reviewed by a hard-core CFO, sophisticated investor or similar kind of expert. And in the case of defining mobile/web/software, there is even more variability in terms of form and format. So, I promised this founder to do a post talking about how you go about create specifications of your MVP.
Before you begin reading all of the stuff below, it’s going to be new, different, confusing. You likely are writing your first one of these. Don’t stress over format. Don’t stress if you are doing it right. This should be an iterative process with advisors and customers providing feedback on the product. Conversations with a technical advisorsor possible developers should be iterative. In fact, let me provide an important warning:
If you create these documents, don’t have input from a technical resource, take it to a development shop and they provide you a price. Go find a new technical resource.
So, just get down what you can in a form that works for you. I don’t expect you to provide all of these things. Part of what I like to find out is where you are relative to capturing these things.
The key first part of the conversation with a developer is having a good capture of the business more broadly. The business canvas model is a good short list. Or you can have a pitch deck. You might also have a business plan, marketing plan, financials, competitor analysis or other kinds of background document. Ideally you are also able to say what you are really trying to prove to get to the next level. For example, if you are trying to determine viral coefficient (see Startup Metrics), then the focus should be around those aspects of the MVP. It’s important to know where the business is today and what you are really trying to achieve. Quite often these things get old quickly. It’s fine to send out documents that have older information. Just make note of it in the document and/or in an email when you send it. Seeing the evolution of thinking is not a bad thing.
Customer Development Notes
I’m assuming founders are having customer development conversations. It would be great to get notes and summaries from these. See also: 12 Tips for Early Customer Development Interviews, 12 tips for customer development, tips for customer development.
Prioritized User Stories
Product Feature List
Create a prioritized high level product feature list (see Product Backlog). Make sure you look at 32 Questions Developers Should Ask a Startup Founderto spark possible additional features. Make sure you keep focused on your key business drivers and prioritize the features aggressively based on those drivers. Examples:
For the higher priority features, begin to capture additional level of detail of those features particularly focusing on the behavior of the application. See User Story is Worthless – Behavior is What We Need although you don’t need your behavior description to be as formal as what is presented. Really this begins to be a Functional Specification. Developers don’t need everything to be fully documented. Rather are just looking for more detailed description of the features/functions. Examples (these are going to be more detailed than you need to get to at this point):
Wireframes, Comps, Clickable Prototype
Create wireframes for a few key screens that sketch your concept using a tool like Balsamiq. Send across any graphic designs you currently have. If you happen to have a clickable prototype, that’s great. That’s fairly uncommon.
Here are other things you might be told about and try to capture:
- Personas – Examples: http://www.uiaccess.com/accessucd/personas_eg.html;
- Use cases
- Software Requirements
Here are a bunch of additional resources
- What is the minimum viable product?
- Minimum Viable Product: a guide, Lessons Learned, Eric Ries, August 3, 2009
- 10 examples of minimum viable products
- Minimum Viable Product revisited – the MVP Curve?
- Don’t Let the Minimum Win Over the Viable
- Minimum Desirable Product
- How I built my Minimum Viable Product
- The Death of the Wireframe? Towards An Integrated Approach to UX Design
You should definitely look at Steve Blank’s book and review his blog.