Startup failure in the UK is lower than in other countries but still remains high at 60%, according to growthrate.com. This means that almost two-thirds of launched businesses have failed in
the past three years. Therefore, there are plenty of lessons that can be learned by entrepreneurs who are considering or launching their businesses. In 2023,
Wilburlabs conducted a survey among 150 founders to find out the reasons for business failures and made recommendations on how to prevent them: “Running out of money” was the first in the
failure reasons list, while “More research prior the launch” was among the top 5 recommendations.
Hypotheses
As Steve Blank says at the beginning, when startups come up with their business model, what they have in reality is a number of hypotheses or guesses. But how do they happen in the first place? Understanding
of the market is a starting point. Hypotheses should contain some already known information, something that is yet to be found out, and use the following formula:
Action + impact on qualitative indicator + impact on quantitative indicator (optional)
A good hypothesis should consider customer needs, trends, context, and the founders’ current resources. At an early stage, a startup might have not one or two, but many hypotheses. How to pick those that should be verified in the first place? One of the
possible ways for prioritisation is a money-making opportunity.
Research, research, research.
Given that you came up with a short-list of hypotheses or guesses and you want to make sure that they are right before you rely on them, you need market research. It helps to sieve out or adjust the wrong ideas, back up the right ones with data, and make
a clear prioritisation of what should be included in an MVP (Most Valuable Product).
Unfortunately, there is no “one size fits all” solution for market research as everything depends on the hypothesis that dictates the goal of the research and therefore conduction means.
Anyway, it’s always good to check whether there is any free and easy way to verify your hypotheses: this may be any statistics, reports, or even a rapid prototype. If there is none, then quantitative and qualitative research are at your service.
Quantitative research can be very helpful for measuring possible demand. The experiment itself should not involve too many resources: it can be a survey, sending our emails to see how many of them got responses, creation of a simple landing page with product
details, and how to reach out to you.
Qualitative research needs much more cost and effort, but it is unavoidable, as it is where you can assess whether your hypothesis about customer needs matches the reality. One of the most common tools here is interviews.
Design your interviews
To get the most from the interview process, you must plan it very carefully. Always bear in mind the goal of the research: what
information you want to get as a result, and how you want to use it afterwards. You also must define what types of users are the focus of your research: the ones that use your products, ones of your competitors, or even none (edge-case
users).
Compose an interview guide with the list of hypotheses, goals, introduction, and a questionnaire. Break up the questions into subcategories. One way of doing it is by a sequence of possible actions on the timeline. Plan your questions for each subcategory to
move from general questions about the topic toward topic-specific and wrap up with clarifications and confirmation questions: lead the conversation to the required depth. If you choose to skip the first part and get straight to specifics, you are more likely
to get a surface answer.
The form of the question and the way you ask it are also important. There are several things to consider:
- ask only one question at a time. People tend to remember the first and last things they hear, so the middle part is likely to be left unnoticed.
- ask to tell the story. When a person tells a story, they automatically highlight the most important things for them.
- ask open questions unless you want to receive a confirmation of an idea.
- ask about experience, not about a hypothetical situation. People tend to believe that they can forecast how they will act in a certain situation, but the truth is, they don’t know.
- avoid terms that might be unfamiliar to your respondents. People hesitate to clarify, and you will receive a surface answer or “don’t know”.
- avoid “Why” questions. People tend to make rational explanations even if they don’t know the answer.
The last thing to do after all questions are written is to check whether they cover all research goals.
Unfortunately, there is no fixed number for the recommended number of interviews, however, the rule of thumb is to run as many as needed for respondents’ insights to start repeating.
Another thing to consider after you have a guide is where to get people you’re interested in.
There are several channels that can be your sources: internal (client managers, sales, surveys, emails, client databases), dedicated (recruiters, chats, communities), and personal networks (social network applications, professional and personal networks).
Do not forget to budget rewards for respondents. Rewards can be in the form of discounts or gift vouchers.
Plenty of data points. Now what?
First thing to do here is structure your data. There is a range of available frameworks that can help: Customer Journey Map, Experience Map, Service Blueprint, Jobs to be done, Portraits, or a mixture of them. Once it’s done, verify your hypotheses with
the collected data, alter or reject. Get the core, prioritise it using quantitative research results, decide what will be included in the MVP. Build a prototype (Wire Frames, UX) and conduct solution research to confirm with the target audience that all core
features are included, adjust the prototype based on the feedback. Now you are ready for the implementation of your MVP in flash.
It’s natural to feel the urge to launch and start offering your service immediately. However, a startup thrives on continuous building and pivoting, adapting based on feedback to achieve the right product-market fit. The more thorough your primary market
research is, the fewer pivots you must make in the future, and there are higher chances that you won’t run out of cash before your business becomes profitable.
