Customer Development 101: Cust Dev for Product Managers in 10 minutes
A short summary of key ideas, concepts and patterns about customer development.
What is Customer Development?
Customer development is a methodology for building businesses that is applicable to startups and existing companies. It is one of three pillars required for a lean startup:
- customer development;
- business model design (use of lean canvas, business model canvas, value proposition canvas, etc. to define and adapt business model and get product-market fit);
- agile engineering (self-organizing and cross-functional agile dev teams that deliver value incrementally, iteratively, and adaptively).
In brief words, customer development as part of a lean startup focuses on understanding customer problems and needs, defining a repeatable and scalable sales model for your product/company, and helps to get product/market fit.
4 Phases of Customer Development:
Each of these phases can take a few iterations before getting the proper results and getting product-market fit. And it is totally ok since quite often initial hypotheses about users/needs/product/traffic sources may be wrong or require significant tweaks — “No business plan survives the first contact with customers”.
- Customer Discovery— answers two questions: (1) who your customers for your product are? and (2) is the problem you’re solving important to them (what are their actual needs)? Customer discovery requires a lot of user research work.
- Customer Validation — is about building a sales process that can be repeated by a sales and marketing team and can be ROAS+ profitable. It answers the question, whether you can actually sell the product (or its MVP) to early customers for real money.
If you succeed in the phases of customer discovery and customer validation, you have proven your business model. If not, you need to pivot and test other hypotheses.
- Customer Creation — is all about scaling customers' volume to your product and increasing demand for your product. Adding new traffic sources, expanding user segments, creating promotions and referral programs — all are aimed to get new customers.
- Company Building — establishing more formalized structure, processes, and organizational departments, hiring more employees to create new procedures, scale, and support existing processes.
Getting through the first two phases - customer discovery and customer validation — guide you to find product/market fit, where most start-up companies and products fail to reach.
Customer discovery is the core part of Customer development.
Great products came from great ideas. Sources for new product ideas can be internal and external to your company. Product managers need to be able to build a systematic process for generating, analyzing, and prioritizing ideas that come from different sources and stakeholders.
All products are created either to:
- solve a problem;
- address a need;
- create joy and entertain.
One of the most effective ways to validate your ideas and hypotheses is a Customer Discovery via user research.
Nothing is more important than user research. Companies often assume they understand their user so well that they can build a product without talking to their users at all, and they’re almost always wrong.
Without performing user research, you’re betting on your ability to totally guess at what a user needs and you’re betting that you can build a product, put it out into the world, and people will flock to it.
So, to avoid that and give yourself the best shot possible, you’ll spend a lot of time looking through secondary research and conducting primary research with surveys and customer interviews.
Remember, a product starts as just an idea quite often based on assumptions. Through customer discovery, you’re trying to turn your assumptions about the customer, the market opportunity, and the business model into proven facts.
To run the customer discovery phase, you need to ask yourself:
- Have we identified a problem that customers want resolving?
- Does our product solve this problem?
- Do we have a viable business model?
- Have we learned enough to go to the market and sell our product?
Answering these questions requires research from both secondary and primary sources.
If you’re worried that talking to potential customers will reveal bad news about your idea, then it’s probably a bad idea.
Target Markets and Target Users
Products are created for users, but not everyone is your target user. You need to talk with people that actually may find value in your product (product for US lawyers may not be interesting for US truck drivers).
You may start with Demographic information to describe your target users initially by:
- job title
- marital status
When you start talking to users you may notice some additional patterns that give more insights into your true target users’ attitudes and other psychological criteria. These are called Psychographic information. People who are vegetarians or believe in a flat Earth are psychological profiles, for example.
Psychographics is a better way to segment a target audience than demographics. It expands your potential target audience above age or demographics brackets. For example, a 20-year-old woman can have similar interests as a 48-year-old man and both may fit our target audience based on a psychographic profile.
You may need a Screener to filter users you want to talk to later. It is based on questions that help you to define your assumed target audience. For example, you may ask the user’s age, gender, job position, interests, etc. Use digital tools like Google forms, Survey Monkey, or Typeform to create a digital screener survey to collect and process users’ responses.
Your survey needs to be disposed to a potential target audience. Not relevant responses have no value for you. So, you need to think, where to find your target audience: posting in thematic FB or LinkedIn groups, website topic forums (Reddit, as one of many), or launching targeted ads. Later these traffic channels may be used again for Customer validation with your product MVP.
Also, you need to think about how to make users interested in taking your Screener survey (offer them something in return — valuable content, promise to give free access to a future product, etc.)
Secondary sources include anything that you don’t learn directly from your users — these are market research, scientific papers, experts publications, interviews with experts. You may want to check these sources initially to quickly learn if your initial assumption is in the right direction. And this is good to know now before wasting any more time.
As a start-up, you don’t have a lot of funds to invest in user research, and you need to aim to get maximum info for the fewer expenses. So you can benefit from free internet sources and researches done by others that are freely available on the Internet:
- just Google it
- look on Google Scholar articles
- check info from private and public research institutions (for example Pew Research Center) on your subject;
- check statistical resources for information (such as Statista) or annual digital reports from market leaders, industry magazines, news sites (for example, We Are Social digital report, State of Mobile App Annie report);
- use analytical spy tools for your niche (for example for mobile apps, you can spy on your competitors and learn how many users they have and money they make via market analytics App Annie, App Magic, Sensor Tower, etc.)
- watch interviews with subject experts or talk to them (they already invested hundreds of hours of user research in your area, so they can help to improve your hypothesis and give insights)
You’ll use this information to augment your primary user research, but all information at this stage is valuable. In some cases, a start-up may stop on secondary sources and do not invest enough effort in actual talk with users. This approach may work if you plan to launch a product on a well-known market, you copy market leaders and know what features are valuable for users, and you can develop and iterate your MVP fast based users’ feedback. But even in this case, it is recommended to talk with the user before customer validation just to double-check that your assumptions are correct.
In a startup, no facts exist inside the building, only opinions, so it is important to “go outside” and talk to some real users. Talking directly to your target user is called primary research.
User interviews and user surveys are two primary methods for collecting users feedback. Both depend on the quality of your questions.
They serve different purposes:
- Surveys help to gather quantitative feedback (for example to measure users' attitudes, define preferences among suggest variants but are not good in learning underlying user needs or desires). It is not recommended to put open-ended questions in user surveys (since users may skip them), it is better to leave them for an interview where you can ask for more details with clarifying questions. A good survey helps to measure how many users feel a certain way about something that you believe in.
- Interviews help to gather qualitative information. When a user speaks in front of you, you can get more insights by observing the user’s non-verbal signs, tailor conversation in the right direction, ask Whys questions to get to deeper needs and motivations. A good interview helps to clarify who your real customers are and whether or not the problem your product solves is important to them.
Interviews are flexible and effective tools for gathering user feedback. You need to make an interview comfortable and enjoyable for your user — build rapport with users, think of them as friends, use your body language to make this conversation more comfortable for both. Show curiosity and respect to user’s opinions (even if you disagree). Your users should talk more than you. Remember, when you are talking, you are not learning.
Tips for interview:
- Use open-ended questions (let your user talk and share his thoughts, do not limit user answers with yes/any or provided variants)
- Not use leading questions (don’t fall to bias to build your questions as already having the desired answer in it). Your users may lie to you and say the answer you want to hear.
- Use classic who, what, when, where, why, and how questions.
- Ask for examples or explanations to better understand the user’s thoughts.
- Use five whys questions to understand better the user’s needs and pains.
- Build your interview structure (have questions template with a defined goal for an interview) and adapt it based on actual conversation flow. Start with easy questions, have small talk — to make a user feel more comfortable. After, move to more intense and complex questions. End your interview with some easier questions, so you do not end the interview abruptly.
- This is an interview centered on discovery. Do not pitch your idea or product to the user at this interview, since the user may say what you want to hear instead of what he/she really thinks. If user responses confirm your hypotheses, ask if you can reach them later for a follow-up interview to validate your product.
- Make notes, or even ask permission for recording. It helps you to remember and structure your findings later.
- There are many free interview templates available on the Internet, don’t be shy — use and adapt them for your needs.
- Look for patterns and different perspectives when analyzing responses from interviewed users. Avoid jumping to conclusions till you conduct enough interviews (it is defined individually for each case and how narrow is your segment, but for me, there is the magic number of at least 10 interviews).
Surveys are for quantity information. They can be misused instead of user interviews since are made quickly and cheaply to collect information.
Surveys are great for three things:
- to measure attitude, intent, or task success (how does the user like the current process, how is it important for users from 0 to 10, what is the NPS for your product?)
- to track changes over time (what was users’ attitude in the past, and how it changed after the feature release)
- to track problems (you discovered the issue in your software, but how many users are experiencing a problem?)
Surveys are bad for:
- to discover reasons behind and what users really care about
- to understand actual users behaviors
It is better to use interviews for problem discovery and product discovery (yeah, these are two different interviews). If you interviewed few target users and discovered a problem they face, you may create a survey with answer options to gather quantitative information about whether other users face similar problems, how often, how important those are for them, etc.
You should keep your surveys short (make it possible to take within a few minutes), since large surveys may scare potential users. Every question must have a specific goal or collect quantity insights, questions should be grouped and have logical connections (do not do jumps forth and back, since it is confusing). Ask broad questions first before moving to more specific ones.
Do not use leading questions!!! Questions like, would you use a product like this, lead to biased responses. Also, questions shaped as “do you agree with a statement” lead to a Yes/No answer, but it is better to ask “how the user feels about something on a scale from 0 to 10”. It gives more granularity to users’ responses. Avoid open-ended questions in surveys (leave those for interviews), use quantifiable questions, like rating scales.
When you analyze survey results remember about statistical significance — if you have more replies from target users, then the more accurate your survey is for a target market.
Also, don’t look only at average values of survey responses, look at details for each score given — there may be additional segments of target users hidden, that can be valuable for you and need to be interviewed.
When to Pivot?
Based on your research results, after you talked with users, your hypotheses will fit one of three groups:
- you were right, and your hypotheses are validated with real users >>> you can move to the Customer Validation stage;
- you were partially right/wrong and there are some key differences to consider that were discovered via user research >>> you need to adjust your hypothesis accordingly and it is better to have a few more interviews to make sure you are right this time and move after to Customer Validation stage;
- you were totally wrong and your hypotheses about users problems are not problems at all >>> you need to form new assumptions and hypotheses that need to pass Customer Discovery before moving to Customer Validation (getting first paying users with your MVP or even pre-order form)
If Customer Validation is not successful then you need to do a pivot and form new problem hypotheses and talk more to potential users. For example, there is a problem identified, but no one wants to pay to solve it.
Pivot at early stages is a good thing — it saves your money from being wasted and saves time that you can spend on discovering and developing a product that people really need or will love.
Recommended books on Customer Development:
- “The Mom Test: How to talk to customers & learn if your business is a good idea when everyone is lying to you” by Rob Fitzpatrick;
- “The Four Steps to the Epiphany: Successful Strategies for Startups That Win” by Steve Blank;
- “The Lean Startup: How Today’s Entrepreneurs Use Continuous Innovation to Create Radically Successful Businesses” by Eric Ries;
- “Lean Customer Development: Building Products Your Customers Will Buy” by Cindy Alvarez;
- “The Power of Habit: Why We Do What We Do in Life and Business”
by Charles Duhigg;
- “Intercom on Jobs-to-be-Done” by Intercom team.