Startups are organizations of extreme uncertainty. The first time you write down your business model/plan, it is a document of guesses. Unfortunately, many founders jump straight into building this first best guess. Sometimes, they are very right. Usually, they are very wrong.
I believe that most founders jump straight in because they are unaware of alternatives. The truth is that it is very easy to create hypotheses and validate them through fast and inexpensive experiments. You don’t need to achieve a 95% confidence, but positive indicators help reduce the risk of the more time consuming and expensive subsequent activities. Often, founders will find new ideas and opportunities to pursue that indicate higher likelihood of success.
The three things that must be validated are: problem, solution, and pricing.
The problem is the very first place to start. Too often, founders fall in love with their solution when they should be falling in love with the problem. The best way to validate a problem is to talk to your potential customers. Another alternative is to create a Google Form survey and send it to your contacts, Facebook and LinkedIn groups (if allowed), Reddit and other forums. You can realistically set a budget of less than $200 and 1 week to find this answer. It will save countless tens of thousands and months or years of pain and agony. Ask open ended questions – don’t lead them – in order to discover what their biggest problem is, what it would mean to them if it were solved, and how difficult has it been to find a solution? You’re looking for a strong consensus. You’re trying to solve the number one most felt problem. A pass/fail exit criteria for these interviews may be that greater than 60% of problems, or similar, are shared by the “difficult to solve” responses. That is the problem that you should falling in love with.
Once you have identified or confirmed that you’re working with solving the biggest problem, you need to brainstorm potential solutions. I like to propose at least 2 and a maximum of 4. This should be easy to brainstorm amongst stakeholders – there are no bad ideas at this stage. Once you have generated a few product ideas, test them! Create landing pages for a product that does not yet exist. There are sites such as squarespace, unbounce, etc. that allow you to select a simple template and list the intended product description and features. Create simple targeted ads, set a small budget (a couple of hundred dollars), and one week of concurrent experimentation. At the end of the week, you should have identified which solution resonated with the potential customers the best – which resulted in the most ad clicks? which had the highest conversion ratio (filled out a “more info” form)? Again, you don’t need to have a 95% confidence ratio, but you should set pass/fail exit criteria prior to running the experiments. If one of your solutions indicates strong market desire – you have a winner. If not, brainstorm other ideas and try again. If you’ve identified a serious problem, and a desired product, you’re almost there!
The last thing that you need to validate is price. In the previous experiments you should be able to begin to shape a CAC (Customer Acquisition Cost) based on how many conversions you received from your investment. Was it 1 conversion for $500, or 20? This will help you build your pricing model. With the help of your engineering partners, you should be able to estimate cost to build a solution, maintain the solution, and to acquire customers (rough estimates). You’ll need to turn a profit. Some solutions may be highly desirable, but infeasible to build or cost prohibitive for your market. It’s important to ensure that you have a desirable product that people are willing to pay for. Similar to the previous experiments, you can fake the ability to accept payment. For example, you could ask the potential customers to prepay for a year, or put down a deposit to hold their position for early availability. You can disable the final step of payment, or you can refund them immediately and tell them that the product isn’t quite ready for deposits. At this point, you have people voting for your solution with real dollars. Now you’re ready to build your product!!