AWS Startups Blog
Identifying Target Markets (Startup Founder Sales Series, Part 3)
Architecture can be a highly polarizing subject. For example, I truly believe that I lived in the ugliest building on the planet during my first year of college. The structure was called Warren Towers and it was a blight upon the Boston skyline. Not only was the design hideous, but when held in contrast with the surrounding rows of brownstone, the older college buildings, and the picturesque Charles River, it felt out of place with its surroundings.
The same feeling could be said about startups that sell to the wrong people. I was an investor to one startup in the mobile sales presentation market. The first market they entered, medical devices, was a good fit since their sales people were often on the road visiting doctors and needed an app to manage and present highly technical content. Then they wanted to go wide. They were canvassing construction shows, plastics trade shows, aeronautical conferences, boating expos, and every other industry type event to generate awareness and sales. One year later, they pivoted back to the medical device industry.
There are times when your market is pretty clear. If you are building an app to monitor the movement of cows, there is little doubt about the industry or customers. Many solutions, however, are broader either because they solve a horizontal problem or because there are multiple potential use cases and customer types.
Some of these questions should be cleared up by going through the customer development process. This is a core practice of the “Lean Startup” concept we mentioned at the beginning of this series. When interviewing potential users of your solution, it helps to identify key elements of your Ideal Customer Profile (ICP), or the type of customer most likely to value and buy your product. Some of these elements include:
- Company size
- Company industry
- Company culture
- Level of regulatory control
- Growth rate
- Hiring rate
- Market growth
- Business model
These are a few common examples, but each startup will have their own unique ways of defining their ICP. Perhaps certain technical platforms or capabilities are needed, or adoption of certain practices or processes are required to be a good fit as a customer. For instance, if you are selling a DevOps tool, a core requirement for your ICP would be that a customer have an in-house engineering team.
Founders often think customer development only applies to the process of designing and building an MVP. In fact, customer development is a continuous loop of gathering feedback and validating assumptions about customers and markets. The process of defining your initial ICP is therefore not too dissimilar to refining, modifying, and extending the initial model. Much like the process of building an MVP is an iterative process, so is defining the ICP.
There are numerous tools and mechanisms used in the Lean Startup methodology for helping founders identify the ICP. One method is identifying cohorts of similar profiles and reaching out with a personal invite to ask a few questions. A natural starting place to find profiles is on LinkedIn, but use the communities that make the most sense for your solution and audience. For example, developers are often going to be on GitHub and Twitter, designers can be found on Dribbble, and many types of community can be found on Reddit.
Most people will respond positively if you create a thoughtful message. Not everyone will be able to help, but this process only requires a couple handfuls of people to answer questions to confirm whether or not they see a large enough value in your solution or feature. This same method can be applied to confirming pricing, assessing and clarifying key messaging, and other important questions about the product and market.
With an initial set of findings, you can either confirm or cancel out a proposed ICP. Could a small data set be a false positive or negative? Yes, but at this early stage, you are looking for an ICP where the data demonstrates a significant level of affirmation, rather than to focus too much on edge cases. There is always opportunity to reexamine at a later stage when there is a sales and marketing team in place to investigate further.
One last thing to point out is that many posts about creating ICPs will emphasize creating an ICP before defining buying personas. The reasoning is that a broader understanding of the market is needed before one can define the types of people involved in the sales process. From our experience though, either the ICP is already very well-understood from the onset (like our AgriTech example from before) or the differences in ICPs being explored is not vastly different.
We reverse that order and identify motivations and buyer personas first. The reasoning is that starting with broad motivations leads to a better and more meaningful understanding of the challenges and opportunities your solution addresses. Even though it is called B2B sales, we are not selling to companies. We are selling to and engaging with people and what they most care about (the WIIFM from the last post).
In the next post, we explore the topic of sales messaging and what it takes to create copy that starts sales conversations with the buyers in your ICP.