Solving the Right Problem (ch. 7)

Startups fail for many reasons. Most startups that fail in their first year or two, fail because they have not found a problem worth solving.

Sometimes they have the problem, but find that not enough other people have it. Sometimes they are seeing a problem where there isn’t one. Or the problem isn’t painful enough for customers to want to switch solutions. Or maybe the real painful problem is in some adjacent market segment.

To hit product market fit, you have to be solving a real problem for real customers. You need to confirm you have a real problem to solve. Then you need to confirm your solution is worth paying for, repeatedly.

You can sense early on that you are solving a real problem. As a human, you’ll recognize positive feedback when you see it. Sometimes just telling the customer about your product makes them to pull out their credit card. Watching the delight on their face as they use your product signals that you are onto something. The feeling that you are solving a real problem is priceless. It’s a significant part of why many of us become entrepreneurs in the first place. To solve problems for people!

Positive qualitative feedback can mask the economic realities of surviving as a startup. In what follows, we are going to rely on quantitative measures. You need data to know you’re solving a real problem. You need data to optimize your product team’s development speed. And you need data to improve the quality of your product so it is hard for competitors to catch up.

Testing the Problem

Start by conducting “problem interviews” with target customers. You should do this if you are just getting started. But you should also do this if you have a product built already, and you haven’t done problem interviews.

The two best resources on customer interviews are the book Talking to Humans and the Grasshopper Herder blog. For now, I’ll give you the basics.

The main rule of problem interviews is do not talk about your solution. I am not kidding. This isn’t a sales meeting, and it’s not a solution or usability test. This is a study to verify your target customer has the problem you think they do. You are there to listen and take notes.

1. Make a simple persona for your target customer.

There are about a bajillion articles on customer personas on the Internet. But my favorite is by Janice Fraser and nicely summarized by Tristan Kromer. It is a simple 4-up, one-pager with a sketch, demographics, behaviors, and needs/goals.

2. Make a list of potential customer problems for your interview script.

Using the needs/goals section of the persona, to derive 3 to 5 problem statements. It’s best if these are worded as “I” statements from the customers’ perspective. List them on a sheet of paper that you bring with you to the interviews. Next to each problem statement, write Yes/No, and a scale of 1 through 5.

3. Interview your target customers about their problems, one at a time.

Find 10 to 20 people like your persona and invite them to coffee. If possible, bring your friend/partner/assistant to take notes for you. That way you can focus on conducting the interview and on listening carefully.

After some initial pleasantries, ask them about each problem, one problem at at time. You want to know if they have the problem or not. If they do, circle the “Yes” next to the problem. You also want to know, if they do have the problem, how painful it is to them. Have them answer on a numeric scale, 1 for not a big deal, and 5 for extremely frustrating.

3. Before completing the interview, ask them if they have any other problems in this area you didn’t ask about. This last bit can provide unexpectedly awesome product ideas. Leave some space on your script where you can fill these new problems in.

4. Interpret the results.

After a dozen or so of these interviews, you should start to see some clear trends. But wait until you have all 20 interviews done before drawing conclusions. If you’re having trouble finding 20 people who are your target customer, your persona is probably wrong.

Interpret the results by adding up all of the severity scores for each problem statement. Use 0 for “No”, of course. You should have each problem ranked on a scale of 0 to 80 (20 answers with a maximum score of 5 each). Sort them from lowest to highest in that order.

Any problem with a total score under 30 is a waste of time. Any problem over 75 should have your full attention. Anything between 30 and 50 is probably just a nice to have. Those can be good for solidifying competitive advantage once you have soundly solved the most painful problems, 50 to 80.

The data should tell you whether or not you are solving a real problem. If you already have a product, you have hopefully focused your product on the problem that is most painful for most of your customers. If not, it may be time to pivot.

Great! So you have discovered a problem that is actually worth solving. And your initial product is a solution that customers appreciate enough to pay for. You know this because you have already succeed in charging money for your product. Right?

Testing the Solution

How do you know your solution solves the most painful problems effectively. Let’s start by setting some target metrics that show we’re on the right track.

Your average *customer acquisition cost* (CAC) should be at about one third of your average customer’s *lifetime value* (LTV). There is quite a lot of meaning packed into this deceptively simple statement. So, let’s unpack it a bit.

There is plenty on the Internet (and in books, if you still read those) about CAC and LTV. So, I am not going to go into depth here. Bluntly, you will have to spend money on customer acquisition. It might be through advertising, content marketing, or direct sales. The cost of these activities, divided by the number of customers you get, will add up to your CAC.

Customers will pay for your solution. It may be a one-time sales transaction, or a recurring revenue model like a subscription. You want customers who buy once to return to buy from you again. This recurring activity is called *retention*.

Some will stop being your customers after a certain period of time, a concept called *churn*. Churn is the inverse of retention. Perhaps you have solved their problem for good, or they no longer have the problem, or they (gasp!) found a better solution. In any case, the average amount of time they stick around is their *customer lifetime*. The amount of money they spent on your service or product during their lifetime is roughly their LTV.

Successful customer acquisition is more validation that you have identified a real problem. Retention of those customers is validation that your solution actually solves that problem.

To survive, you will need to acquire an increasing number of customers. And you will need those customers to buy more of your product or service, more often. Money earned through sales must be reinvested in building, delivering, and supporting the product. But it must also contribute to more customer acquisition.

Your revenue and retention should increase faster than your customer acquisition costs. Maybe your customer acquisition become more efficient as you learn to improve marketing. Maybe your revenue per customer increases because the product quality entices them to buy more or pay more per purchase. If you’re not increasing LTV faster than CAC, your startup is in trouble.

Ideally, customers are converting on your site at an increasing rate. They are actively and enthusiastically using your product. And they promote it freely among their friends and coworkers.

So, let’s say you are solving the right problem. Customer acquisition is not your only cost. You also have to develop the product, adding and improving capabilities, and quickly. After all, if you really have found a problem worth solving, it is quite likely others have too.

The competition is coming. You need to move faster, or they will eat your lunch.

Enjoyed this chapter? Get the whole book on Amazon.

Originally published at www.startuppatternsbook.com on January 18, 2016.