The Mom Test: how to talk to customers and read the answers

The Mom Test is a way of interviewing potential customers so that their politeness can't mislead you. Instead of asking for opinions of your idea, you get people to describe the most recent time the problem came up and what they did about it. Rob Fitzpatrick set out the method in his book of the same name.

Most summaries stop at the book's three rules. This guide covers what comes after them for a founder with a physical product to sell online: finding people to talk to, running the conversation and deciding what the answers mean. For a list of questions to ask, see validation questions to ask customers.

What is the Mom Test?

The name comes from the person least likely to criticize your plans. Your mom will tell you the idea is great because she wants you to be happy, and strangers on a call are kinder than you'd expect too. Fitzpatrick's answer, in The Mom Test, is to talk about things that have already happened in the person's life, which gives them nothing to be kind about.

Summarized in our words, the rules are:

• Keep the conversation on the other person's life, and leave your idea out of it.
• Ask what happened on particular occasions, because what someone usually does or plans to do is a guess.
• Let them talk, and keep your own part short.

Products people buy for pleasure work a little differently. A candle or a jacket rarely solves a problem, so if you go looking for frustration you may find none and wrongly decide nobody wants it. Have them walk you through a recent purchase in the category instead, from what set it off to what they paid and what they nearly bought.

A worked example: a no-splash dog bowl

This example is invented. A founder plans a heavy, wide water bowl that stops big dogs from splashing water across the floor. Their first guess at a customer is someone with a large dog in an apartment with hard floors. The interviews need to show whether those people do anything about the mess today.

1. Before the first call, list what you need to know

Write down the handful of things you need to find out, phrased so people can answer from memory. Tie each one to a choice you have coming up, such as who to sell to first or what to charge.

The dog-bowl founder wrote down three:

• Have they done anything about water on the floor in the last six months?
• What have they spent on it, and on what?
• What made them act when they did?

Questions like "Would you buy my bowl?" or "What would you pay?" don't belong on the list. People answer them to be nice, and the answers predict little.

2. Find people to talk to

People you know are the quickest to reach, but they are also the most likely to flatter you. Their friends are a good middle ground, so ask everyone you talk to who else they know. Online communities, local Facebook groups, subreddits and in-person spots such as a dog park are the next place to look. Each group has its own rules on research posts and direct messages, and a post that breaks them can get you banned from the group you need most. People also reply more readily if you've been helpful in the group before you ask for anything.

Keep the message short and honest. Say where you came across them, that you're starting a business in this area and would like to learn how they handle the problem, and ask for 20 minutes. Leave out links and anything that reads like a sales pitch.

Expect most messages to strangers to get no answer, and send more than seems necessary. In the example, the founder messaged 10 people they knew and 26 strangers who had posted about big dogs in two local Facebook groups and a subreddit. Seven of the 10 replied and four agreed to talk. Six of the 26 replied, and four of those talked. That's 36 messages for 8 conversations.

3. Spend most of the call on one story

The core of a good interview is one recent occasion, told in the order it happened: "When did your dog last splash water across the floor? What happened next?" Let the person tell it, then go back over it with follow-ups that cover your list from step 1: "What did you try? What did that cost? Why that one?"

Around that story, you need a little set-up and a little wrap-up. At the start, say who you are and that you're there to listen, give them a time, and ask to record. A few questions about the dog, the home and the floors tell you whether this person fits the group you're testing. At the end, ask who else they think you should talk to and whether they'd like to hear when the product is ready.

Talk about the idea only if they ask, and only at the end. Treat whatever they say about it as opinion. If you're wary of sharing it at all, read whether it's safe to share your idea.

On recording, get a clear yes from everyone on the call before you start. Federal law lets a person who is part of a conversation record it, but some states require the consent of everyone in it. In California, Penal Code § 632 makes it a crime to record a confidential conversation without all parties' consent. Asking each person at the start meets that stricter rule too.

4. Write it up while it's fresh

Put each conversation on its own page the same day. Record whether the person fits the group, the story in a few lines, any money they spent and a phrase or two worth remembering. Only put words in quotation marks if the person said them. Keep what they gave you, such as a waitlist sign-up or an introduction, apart from what they thought of the idea.

Then review the call itself. Draper can read your notes after each call and suggest what to change for the next one.

5. Sort what you heard by what it cost them

Kind words cost nothing, so they tell you little. Fitzpatrick's test is commitment and advancement: did the person give something up for you, or agree to a concrete next step? For a founder who plans to sell on pre-order, the dog-bowl answers sort like this:

A friend's praise is worth even less than a stranger's, but a friend's actual purchases are as good as anyone's. And people outside your group can only tell you about themselves. In the example, a beagle owner loved the idea, which says nothing about owners of large dogs.

6. Read the answers together

Keep booking calls until new ones stop surprising you. A 2006 study that tracked 60 interviews found the main themes had all appeared within the first twelve, with basic elements of them present in the first six. For a first product, that is enough to choose a direction, and pre-orders will test it later.

Then put the people who fit side by side and compare what each of them has done about the need. Write each finding as a count, such as "4 of 6", so you can see how many people it rests on.

In the example, 6 of the 8 people fit the group. Four of those six had spent money on the mess in the last six months: $35, $60, $48 and $85, an average of $57. All four rented, and each gave the same reason for acting, which was worry about losing their security deposit over a damaged floor. The two who owned their condos wiped up with a towel and had bought nothing. Across the six, four joined the waitlist, two introduced another dog owner and one offered to pay in advance. The founder narrowed the group to renters with large dogs, a change that rests on four people. Whether those renters will pay is the next question, and testing demand with pre-orders answers it.

When interviews need a different approach

• If people buy on impulse or as gifts, what they did before tells you less. Ask about the last gift or treat they bought in the category, and move to a demand test sooner.
• If a week of outreach finds nobody who fits, the group may be too small or too hard to reach. Check whether your target market is real before you send more messages.
• If you already have pre-orders, talk to the people who placed them. Their reasons for buying are better evidence than any stranger's account.

Related questions

Once you know who your first customers are, product differentiation covers finding the gap your competitors leave.