• Not My Job
  • Posts
  • How to do customer research when you’re scared to talk to people

How to do customer research when you’re scared to talk to people

The clearest guide on actually understanding what people want to buy

Abhishek has probably spoken with 40,000+ people for customer research and done more than 1,750 customer interviews in the last 2 years.

No, i’m not exaggerating.

So when i tell you that he knows how to talk to people, he really really knows how to talk to people.


Abhishek runs customer research for us at Ignore No More (he was actually employee #2!) and now owns his own research agency Deep Research.


I talk a lot about customer research, customer interviews, customer data (i think you get the picture) and why you need them.


But most resources are just too general to be of good use, especially if you’ve never done customer research or interviews before.

Much less for a product or service that you’ve started tying to your identity (don’t do that!) and love dearly.

so Abhishek and I wrote this nitty, gritty, exceptionally practical guide so you know

  • Where to start (when you have no idea who your customer is)

  • Where to start (when you have bits and pieces of customers)

  • How to know who to interview (so you’re not talking to people you shouldn’t bother with)

  • How to conduct interviews (including what questions to ask) so you don’t ruin the call

  • figuring out what questions to ask

  • how to ask them and let the conversation flow so you get great insights you can use

  • How to go from customer research to messaging and marketing plan

Yes, this issue is long. But bookmark it and never wonder how to do customer research again.

but first! some definition of terms

i’ve been pretty lackadaisical in my differentiating between the types of customer-related activities, which is doing no one a service (sorry!). makes everything i say less useful and more confusing (sorry!).

Here’s a quick breakdown of the terms before we dive in.

  • customer research: this is everything that encompasses learning more about your customers. If you’ve been following our marketing foundations, it’s #1. This covers

    • getting background knowledge on the customers (ex: one:many research like Reddit research, Forums, Review mining)

    • speaking to customers indirectly (ex: surveys

    • customer interviews: speaking to people 1:1 in calls, video, or in person (if you don’t have a 2-way conversation it doesn’t count as an interview!)

  • customer research database: your process and storage spot for tracking and making use of the qualitativeinformation you have on customers from research and interviews. If you don’t have it centrally located, it doesn’t exist.

  • customer data: this is, broadly speaking, the quantitative information on how customers interact with marketing materials (emails, website) and product, usually split between your marketing tools and your product data software.

  • target audience: the broad swath of people who might be interested in your product or service

  • customers: do they pay you?

  • users: use the product, but don’t pay you in cold, hard, cash

  • ICP’s/personas: the subsection of your target audience (including customers and users!) that has a specific use case, demographic, and/or psychographic need and can be grouped together (what does ICP stand for? even i don’t remember anymore 😅. According to Google, it’s “Ideal Customer Profile”, which is not how i sue the phrase, but tomatO tomAto 🍅)

Where to start
(when you have no idea who your customer is)

When you’re starting from absolute scratch, you don’t really know the problem you’re trying to solve or for whom. You’re going off of ✨vibes✨ and intuition.

Use that to your advantage instead of being

by starting with what you don’t know,
and using a “1:many” conversation approach.

That’s when you’re looking for people talking generally about the problems you want to solve.

But they’re not talking to you in particular about them.

Examples:

  • Reddit

  • Forums

  • Reviews

Here’s how to do it:

  1. Evaluate the features, problems solved with the product, and current suspected use cases for the product (or industry at large if you haven’t got a product idea in mind yet)

You can use ChatGPT to ask for new possible audiences around the core problems solved.

  1. Do keyword research in a tool like Ahrefs, SemRush, or Google Keyword Planner to find out what target-audience-specific search terms have the most traffic around

  2. Do community research (Reddit, forums) to find out what kinds of problems people are having and what ways they’re describing them, and who the bottlenecks are from their perspective

  3. Do review research (also called review mining) to find out how people are speaking about competitors’ solving their problems (or not). this step can be skipped if you’re not at a product stage yet.

once you’ve done this, you should be able to answer: 

  • your niche industry/use case

  • your core target audience/ICP/Buyer persona

  • the words and phrases that your target audience in your use case/niche wants to hear

if you can’t, than you probably need to narrow down your problem a bit based on what you’ve found are possible problems in the industry/a use case, and repeat the steps.

How to know who to interview (when you have existing users)

When you already have customers, you should be starting with talking with them.
Then start talking to more people!

Here’s how to go from “a few customers” to knowing how to find and get more of them.

First, talk to existing users - so you can understand who they are and why they’re using the product. Try to talk to 3-7 people.

Then, test channels and messaging based on that initial ICP.

Reach out to people in communities where they would hang out:

  • LinkedIn,

  • Reddit,

  • Facebook,

  • Private Slack communities

  • Other forums,

  • WhatsApp groups,

  • Telegram groups

  • Discords

  • Totally depends on the buyer persona.

Hive Index is a great place to find communities.

How to find people to interview (only after you’ve narrowed down who you should be speaking to!)

The first 2 weeks you’re testing messaging + channels.
After from there you’ll be able to tell where your people are - and what messages hit.

From there, focus on the channels with the highest response of qualified people, because that’s where you have your message-marketing fit.

If you’re in a specialized industry (ie, accountants, lawyers, therapists, etc) where hourly rates are very common, you will probably have to do recruitment.

AKA finding people with a paid service where you pay per person or group of people.
Or you’re paying people’s hourly rate

That’s often $100-$150/person (but if you need to talk to executives or C-Suite, it can be higher)

Your goal should be to:

  • for new products less than use cases (or you’re on a budgeto) speak with 20 to 25 people

  • if you’re narrowing down between a lot of use cases, or you have the cash, speak to 30 to 50

Speaking to more than 50 people is just redundant.
You’ll have your answers by 50.

A word to the wise: do NOT mix selling and customer research.
These calls are to learn about people. NOT to sell to them.

Break the circle of trust in a research call or interview and you will both skew your answers and potentially ruin the data and leave a bad taste in the interviewees mouth.

Better to be patient now, for successful messaging at scale later.

How to conduct interviews (including what questions to ask) so you don’t waste a call

how do i prep for an interview?

Here’s a cheat sheet in a graphic.
Save to your desktop and pair with box breathing before a call!

What questions do i ask?

It’s hard to say exactly what to ask because it should really depend on who you’re talking to and what information you need to know.

Here’s a quick guide to get you started:

Yes, you should always have some questions in there that have nothing to do about work.

My favorite one is: What is your favorite movie franchise?
Ask it and watch people’s countenance change as they try to compare John Wick and JRR Tolkien’s Lord of the Rings to choose a winner.

These kinds of questions humanize your customers and gives you the space to create insider jokes and lore that turn customers into raving fans who know without question that you’re building for (and care about!) them.

Over time, these snippets will also give you places for crossover posts as well!

For example, one of the companies we did this for had a huge crossover of Matrix fans.

We made several references to Morpheus in a blog post not long after and customers loved it (to the point that they emailed us to tell us that!).

In what order do i ask questions?

Completely depends. but i’ve found the best order is:

  • a personal question to break the ice

  • work and workflow questions

  • community question(s)

  • end with a pop culture question to end on a fun bonding moment

 Great, i did the interviews! Now what?

Here’s how to analyze interviews for the data part of it so you concretely know what customers want.

  1. Create a spreadsheet (Google Sheets or Airtable)

    • each row is an interview

    • each column is a question

  2. Go through each interview and pull out the core answers to the questions.

    • Make a tag for things like “product features”, “use case”, and so you can easily sort and filter by tags

  3. Then turn it into a chart so at a glance you can see the answers to your questions like

    • What’s the most requested feature?

    • What’s the most common use case?

    • What’s the most hated thing about a competitor?

Leave a notes column at the very end for “great lines and phrases”.

I can’t tell you the amount of times we’ve use a hilarious, perfectly fitting line from an interview as the

(if you check out our site, the whole ticker in the middle is pulled straight from the first 15 indiehacker coffee chats i ever had.)

How do I go from customer research to messaging and marketing plan?

Great question!
Would you look at that…

I have a whole presentation on how to do that!
Checkout slides 26-30 in my “How to get out of a Growth Plateau workshop i did for MicroConf last week

And here’s an image to make it crystal clear (if you blow it up, lol).
In short: how do you answer the question behind the question?

Worksheet copy link lives here.
This is Tab 3.

I did it!!
When can i say i’m done with talking to people and move on to building?

here’s the secret…

YOU’RE NEVER DONE!!!
mwahhahahaaha
maniacal laugh maniacal laugh

No but seriously.
While customer research should be ongoing, there’s definitely a point at which you can say,
“i have enough information to make a clear decision”

It’s when you can answer these big questions and have these data points (depending on what your research goal is:

There you have it.
A complete guide to customer research.

Not gonna lie, this issue combines a good 1,398 hours of sharedexperience into ~2,000 words.

If you need to know your customers (and you do) and don’t want to do this yourself, Let’s talk.

Abhishek and I can get you clarity on your customers and what to do next in your marketing in 45 days or less for a flat price. (I know right?!)

Get on our calendar here: https://cal.com/sophia-o-neal/lets-talk



We’ll see you and your inbox next week!

Cheers!

Sophia ⚡👩🏽‍💻 & Abhishek 💙💁🏽‍♂️ & Aelia 🪄🧕🏽

Powered by ludicrous amounts of coffee and cal links.
Feel free to replenish the espresso supply and buy us a coffee*.*

Reply

or to participate.