User interviews training

User interviews training

User interviews are the foundation for most ideas you propose as a growth team. They’re your source of truth for what the customer wants — but you have to do them properly.

Skim this doc before doing Phase 1 so you have a high-level sense of how this will go.

Phase 1: Prep work

Read the U- Users section of the USERS growth framework.

Then, read the answer below:

When it comes to initial user interviews, wouldn't we want to interview the unhappier users, since they'd have more feedback on what we can improve on?

Well, there's some value in interviewing unhappy users too, particularly around improving onboarding & conversion rates, but it's most useful to interview happy users because we're trying to find more people like them.

Happy users use the product in spite of problems with the it, so we can use them as a signal for which features resonate most and which channels they use to hear about us. We can use the learnings to pick the tactics most likely to get more of the type of users we want.

It's more of an uphill battle to use learnings from unhappy users because they're not the ones actually benefiting from the product. We can’t generalize their learnings — at least, not without many grains of salt.

Unhappy users will tell you what needs to change about the product, onboarding flow, etc., though, so you can focus on them after you’ve learned about the happy users to improve conversion.

^^All of this assumes that there are some ideal users. If we don't have any at all, then we're stuck with interviewing unhappy users (which usually means the product has to change).

Phase 2: Learn

At a high-level, we’re trying to figure out:

  • What growth loops might apply to this product
  • What acquisition channels might work (word-of-mouth? content? Twitter? etc.)
  • Magic moments in the product
  • What their value props are
  • What their buyer’s journey looks like

Tactically, you need to do a few things.

Ask about the past and ground your questions in specifics. “What was the last meeting you booked this week?” is better than “How do you usually book meetings?” and 10x better than “How will you book meetings?”

Almost never ask hypotheticals. Why? People are notoriously bad at predicting what they’ll do. By asking what they already did, you’ll learn their actual behavior.

Ask followup questions. It’s more important to tug on interesting threads than to ask “all” the questions on your list. See where the conversation goes.

E.g., “You mentioned [you like dogs] — let me dig into that a little more. [Have you owned a dog before? What type was it? etc.]

Ask “How did you hear about X?” A lot. This doesn’t just apply to your product. Ask about other similar products they use. Dig into why they tried it or why they trusted it.

In particular, if your interviewee heard about you through word-of-mouth, dig into who referred them and how they know that person. There are often interesting hints around acquisition channels. You might also ask for an intro that person. Rack up those intros.

Basically, you’re filling out phase 2 of the growth strategy.

FAQs

When it comes to initial user interviews, wouldn't we want to interview the unhappier users, since they'd have more feedback on what we can improve on?

Well, there's some value in interviewing unhappy users too, particularly around improving onboarding & conversion rates, but it's most useful to interview happy users because we're trying to find more people like them.

Happy users use the product in spite of problems with the it, so we can use them as a signal for which features resonate most and which channels they use to hear about us. We can use the learnings to pick the tactics most likely to get more of the type of users we want.

It's more of an uphill battle to use learnings from unhappy users because they're not the ones actually benefiting from the product. We can’t generalize their learnings — at least, not without many grains of salt.

Unhappy users will tell you what needs to change about the product, onboarding flow, etc., though, so you can focus on them after you’ve learned about the happy users to improve conversion.

^^All of this assumes that there are some ideal users. If we don't have any at all, then we're stuck with interviewing unhappy users (which usually means the product has to change).

How do I push for a call or more info in a situation where I get answers that are too brief? (See email inside.)
From: Michael Boss To: Nicole Employee Cc: Pranav Marketer
Thanks Nicole.
Hi Pranav,
I found PRODUCT because I invested in it. Otherwise it's usually product hunt for me.
Best,
Michael

On Mon, Jan 3 2022 at 12:51 PM, Nicole Employee wrote: Hi Pranav! happy to!

+Michael my boss!

On Mon, Jan 3, 2022 at 11:00 AM Pranav Marketer wrote: Hey Nicole,

Thank you for doing an interview with PRODUCT. You mentioned your boss is on the cutting edge of productivity tools.

Could you introduce me to him?

I would love to talk to him about how he finds and hears about new tools. I believe it would help us grow PRODUCT.

Thank you, Pranav

There's definitely more info he can give here.

Couple thoughts:

1. Spin the pitch (a call or more info) to his benefit. This advice applies to other emails you write, e.g., cold outreach, email marketing, etc.

Some people are sold by gift cards, but $25 in Amazon bucks probably won't sell him; his time is probably worth more than $50/hr.

One idea: Pitch him on how his answers will make his investment perform better because these calls are how PRODUCT unlocks its growth.

You might also pitch how his input makes Nicole's job easier, which ends up making his life easier/saving him time (you can get more specific on how product input saves him time based on your conversation with Nicole).

2. You could also just follow up via email with a few bullet-point questions if you think he's too hard to pitch on doing a call.

How long should user interviews take?

15-20 mins is a good rule of thumb, but leave yourself some buffer at the end — sometimes the most useful parts of a user interview or the free-flowing conversation at the end. We've had some user interviews that were great that went 40 or 50 minutes.

Phase 3: Come up with a questions doc

Come up with 5-15 questions you can use as crutches in your interview. Here’s an example of a good doc — feel free to steal.

Template: User Interview Questions

Phase 4: Schedule a mock interview session with your mentor [optional]

Ping them on Slack to do this if you’re part of our program. It should roughly go:

  1. You interview them (10 minutes) + feedback (5 minutes)
  2. They interview you (10 minutes)
  3. You interview them again (10 minutes) + feedback (5 minutes)

Phase 5: Pull a list of retained users

Retained users mean people who continue to use you, not just who used you once and left.

Don’t wriggle out of this unless you really don’t have retained users. This list should come from something like Mixpanel or Amplitude or your backend. Really good founders will know these folks by name in the early days.

Tactic: find the testimonials on their website and ask your founder for intros to the people who gave them.

Phase 6: Outreach

Ask your product team or account managers to see if they can make a warm intro to customers they’ve already talked with.

If that’s not possible...

  1. Set up a calendly to make it easy for people to book time with you.
  2. Just use Gmail to send invitations from your client email (i.e., yourname@clientname.com). You shouldn’t need anything fancy and you shouldn’t send more than 50 emails in a batch.
  3. You can use the “Message Template” feature or aText to speed up the repetitive parts.
  4. Here’s an outreach template that has worked well in the past:
Subject: [COMPANY NAME] feedback (and gift card)
Hey NAME!
I work with the growth team at [COMPANY NAME] — I saw you’ve been using us for a while and I’d love to interview you about what your experience has been like. This would directly impact the features we build and how we market them.
(I'll send you a $25 Amazon gift card as a thank you.)
Free for 15-20 minutes this week or next? You can book time here [LINK to calendly].
-[YOUR NAME]

An incentive helps get people to respond, but it isn’t always necessary. For some, the incentive is just being able to give input on the product and marketing direction, and adding money/gifts makes the interaction feel too transactional. (We see this most with mission-driven companies.)

Sometimes, we offer a “virtual lunch” to be cute – i.e., a $25 DoorDash gift card. You get the best response rates when you research the person you’re emailing and offer them a tailored incentive. For example, if their Instagram has tons of photos of their pug, get them tickets a local dog show.

💡
Note: If your company has fewer than 10 happy users, you should invest much more than $25 in them. Good teams literally fly out to get lunch with these people and build a relationship with them. (This also makes it easier to interview them in the future.) You can invite these people to your “Customer Advisory Board.”
  1. Bounce your email off your mentor in Slack for feedback.
  2. Use the “Send Later” feature in Gmail to schedule the email to go out between 8:30 and 9:10 AM on a weekday. Account for time zone.
    • Why? People check their email first thing in the morning (i.e., 9 AM, when they first start working), and the newest emails sit at the top. This means we want to get to their inbox right before they open it.
    • There are many opinions on the best time to send emails, so feel free to ignore our advice if you think you have a good reason.

Phase 7: Interview People

You can build off the questions in Phase 2.

It’s OK (in fact, it’s very good!) to ask for another interview if you weren’t able to get to every question you wanted and the vibe is good.

Aim to interview at least 10 users. Stop when you aren’t getting new learnings.

Feel free to invite your advisor to the first 1 or 2 interviews (if it works with their schedule).

[Optional] Record calls with Grain.

FAQs

I don't want to distract users/myself by taking notes during the interview. What do you recommend?

Don't beat yourself up if you don't catch everything. A few ways you can do this:

  • Record the interview if it's on Zoom or use a tool like https://grain.co (ask for permission first) and take notes on it later when you re-watch.
  • Don't take many notes during the call, but write down your learnings afterwards
  • Split your screen. Keep the Zoom call on the left and your notes on the right, and say up front "Is it OK if I take notes while you talk? I want to make sure I totally understand what you're saying and taking notes helps — sorry if my screen shakes!"
Is it OK to ask users if I can record the interview?

Yes, definitely. Just ask up front if it's OK and also explain why you're recording it (e.g., "I'm recording it so I can fully listen and not worry if I missed something you said.")

How do you troubleshoot why users aren’t doing user interviews?

Expect at least a ~40% response rate. Debug by looking at:

Open rates (are they even opening your emails?)
  • Check if you're getting marked as spam (use, e.g., Glock to do this)
  • Improve the subject line + first line of the email to better stand out.
  • Look into trying to get a warmer intro if you’re sending emails cold.
Response/book rates (are they opening and not responding?)
  • Look at the incentive and email copy/pitch. Bounce it off your mentor.
  • The incentive may not be big enough. Try paying more and making it more personalized to get them to talk. E.g., a gift card to a michelin star restaurant. (Or maybe tie it to your research on them, e.g. a gift card to a dog cafe).
  • Remove the incentive entirely — it may make the ask feel too transactional.
  • If you’re reaching out cold, try getting a warmer intro. Can you send a forwardable blurb to the Customer Success person owning that customer?
No-show rates (are they ghosting?)

Look into sending email and text reminders.

Incentive

Is your incentive personalized or just a generic gift card? Buying an opera fan tickets to the next Plácido Domingo performance will hit better than an Amazon gift card.

If none of these work, it’s a signal users may not be as happy as you think.

Phase 8: Create Outputs

Your learnings should lead to...

  1. Filled out personas slides and buyer’s journey
  2. A filled out value props doc (template, background reading)

Start filling them out.

Additional Resources