Until you get specific with an idea, all products seem good.
Don’t ask future hypothetical questions. Ask about specifics in the past.
Talk less; listen more. If you are doing more talking then you are not learning; you are pitching.
A user who doesn’t have time to try the good-enough products that exist today won’t have time to try your perfect product tomorrow either. This user won’t value your product regardless of what they say.
You should charge based on what value you create not what it costs you. See how much the user is paying currently to solve the problem.
User research is about the user and not your product. If they start complimenting you, then you are doing something wrong. If they complimented you then immediately bring the meeting back to the learning conversation.
Users’ ideas and feature requests should be understood but not obeyed.
For unpleasant tasks, ask yourself if you delegated this task to someone else, what would they do? And then do that yourself.
In case of product risk, for example, games, it is better to build the core game-play mechanics and test it early on.
The early user meetings should be quick. It is better to do them spontaneously in person, at say conferences. Video calls and phone calls are expected to be more formal.
During a customer research avoid giving too much information about what you are working on to avoid biasing them. However, it still, nudges the discussion in a useful direction.
Every meeting succeeds or fails. You lost a meeting if it ends with a compliment or a stalling tactic.
To identify real early adopters, look for currencies of commitment. A compliment costs nothing and it is worth nothing. A real commitment is cash, time (e.g. committing to a trial), or reputation.
At the initial stage, the goal is not to get users but to get learning about users. So, adding temporary friction can help identify serious ones.
The rejection rate, of say cold calls, is completely irrelevant at the initial stage. The goal is to find some potential users to learn from and not to build a reproducible sales pipeline.
The ultimate excuse to ask random people is to claim that you are doing Ph.D. research or claiming to write a book.
How to send cold emails (VFWPA – Very Few Wizards Properly Ask!)
Vision – e.g. helping busy teachers in classroom activities
Framing – tell them if you have a product to sell
Weakness – tell them about the product, this shows one has done the basic research
Pedestal – Show how the user is uniquely positioned to help
Ask – ask for the help
When the customer feedback is all over the map, you cannot extract value. You need to narrow down to one specific segment of customers to actually start learning.
The best early adopters are reachable, profitable, and personally rewarding. The last part is really important. Choose customers you like, admire, and enjoy being around. Customer research is draining and if you don’t like the market you are in, you are going to drain yourself out.
You can make three mistakes in early segmentation – picking the wrong segment, overlooking the right segment, or not talking to all the stakeholders in the case of a business.
If you don’t know what you are trying to learn, don’t bother having customer conversations. You need to have some hypothesis of what you are trying to learn first.
Best to have two people doing user research together. One takes notes. One interview. More than two, and it starts to feel like a focus group. And the customer becomes intimidated. If you don’t have a co-founder, find a friend who you can count on.
Customers are surprisingly happy to be recorded. However, almost no one goes through recordings later, so, there isn’t a point in collecting one. Best to transfer notes to a spreadsheet afterward along with an emoji symbolizing emotions.
Do not take notes on the laptop. It makes the conversation more formal. The only way to make it work is to put the laptop on the side and do touch typing while talking to the customer! And of course, tell the customer that you are a super-slow writer!