Hola 👋🏻
As I promised (threatened?) last week, I’m going to dial up the helpfulness of the content I share with you. This week I’m giving you everything you need to do your first bit of research.
This is my tried and tested approach to doing research. You’re getting 26 years insight on how I’ve done research myself, commissioned it in design teams, or watched others doing it. So this is proper good stuff.
Before we go any further. A disclaimer.
Unperfect is a Validation-Free Zone
You may have heard of validating an idea or someone say “let’s validate that”. We don’t do that here.
Validating or validation testing introduces bias into your design process from the start, and that’s not a good thing. Bias in design has led to some truly shocking badly design products in the wild.
Research is used to understand people, products and problems, it can have a rainbow of different outcomes you will learn from it.
Whereas if you do validating or validation testing, you’re only looking for your idea to be valid in the eyes of the people you’re talking to. You become blinkered and any failure is MUCH harder for you to accept.
Open your mind. Try research instead.
Thank you, and with that, let’s crack on.
Preparation
In this week’s beginners guide to doing research, I’m going to take you through the prep you should do before you carry out research. This means I’m going to take you through the following steps:
- Learning objective - what do you want to learn from the research
- Discussion guide - guides vs scripts
- Interview plan - the sequencing of the sessions
- Consent forms - getting permission to carry out research
- Compensation - to compensate or not?
- Recruitment - who, what, why
- Testing your research
The type of research we’re covering today will be customer interviews, in the next section I’ll take you through two different ways you might want to use customer interviews, but it’s not exhaustive (we’d be here all day if I did that).
I’ve also created you a handy little download so you can print it out and work along with the steps.
Unperfect Research Workbook V1290KB ∙ PDF fileDownloadDownload
1. What do you want to learn?
The first thing I always do when doing research is write down what I want to learn. And I don’t mean think it up and crack on. I mean physically writing it down.
The reason why I do this, is to help me identify the right kind of research to use. It also helps to keep me super focused on why I’m doing the research. We call this a constraint, without them you could go off on a tangent and find it hard to get back again. Constraints are amazing when you learn how to use them to your advantage.
I’m giving you two examples of how you might want to create a learning objective for your research. If you’re stuck or need help troubleshooting your objective, slide into the chat and ask, I’ll give you some help over there.
Starting a business
If you’re starting a business understanding the people who will pay you money is the most important thing you can do.
The things you’re going to need to try to find out include
- Who they are
- What they do
- What problems they have (obvs related to the problem you want to solve for them)
- Who they currently get to help them solve their problems
To do this successfully, your objective should look something like:
I want to found out who my ideal user / customer / client is and what problems they have.
When you go with this approach you’re giving it major 🤌🏻 because you’re opening your mind to *all* the possibilities you might be shown.
This is what will happen:
- What you see as a problem might not be a problem (or as big a problem) for the people you want to solve it for.
- It might be a problem but the people you want to solve it for might not want to pay anyone to solve it.
- You might have completely the wrong audience for the problem solving.
- Or, my favourite, you might find an even better problem to solve.
The last one is the absolute belter of a benefit you get when you don’t try to simply validate your idea.
Don’t do this
I know you (probably) have an amazing way to solve the problem of the person you think you want to help. But, if you use a learning objective that looks like this
I want to find out how my solution will solve their problem
You won’t learn very much at all.
The reason why is that you’re only going to be hearing things that confirm your assumption. When they don’t, you’ll get upset or angry at the way the interview turned out. Or worse, the person you want to help.
Launching a new product
So you’ve already got a business, it’s been going for a little while. You have an idea for a new product or service (let’s take a mo to celebrate this, you maven you).
The thing with this is that you probably know your audience really well by this point. So finding out about your customer and the problems they have isn’t going to help you as much here.
Things you’re probably going to want to find out when you do this research:
- Do your parish know it’s a problem for them?
- Do your parish currently have a way to solve this problem?
- How do they solve the problem?
- Why do they need help solving this problem?
- What language are they using to talk about it?
To do this, your objective might look a little something like this
I want to learn more about the specific problem I know my parish has
You may already have anecdotal information that covers the answers to the above questions, but by formalising it with 5 - 10 members of your parish, in a structured way, gives you confidence and certainty.
The language they use to describe the problem you can help them solve might not be the same as the language you use. By understanding how they talk about it makes it much easier for you to meet them in the middle.
2. Discussion guides
When it comes to doing research, you want some questions to ask the people you’re talking to. The reason why I opt for guides over scripts is because a good piece of research is a conversation with someone. There are things you absolutely want to know from each person you talk to, and then there are things you want to explore by letting the conversation take the direction it naturally goes in.
A script will make you sound mechanical and you may not hear what you need to hear from the person you’re talking to. You will also spend more time trying to remember where you are in the guide, finding the next question and fumbling. It will make your life so much easier if you have a guide.
Go wide
To start off with, what I want you to do is get a bunch of post-it notes, real or virtual, or a pad of paper.
Set a timer for 20 minutes.
I now want you to write down every question that comes to mind that you want to know from the people you’re going to talk to.
Do not filter yourself.
When you’ve done that. Put them down, and leave them. At a minimum, give it an hour, but ideally, leave it over night.
Sorting
When you come back to your list of questions, I want you to sort them into logical groups to you. You’ll probably notice that there are questions that relate to one another. This is why post-it notes are best for this.
If you’re stuck, here’s some ideas of how you might sort them:
- Intro’s - learning about the person you’re talking to and what they do
- How’s - learning how they do something, or solve something
- Why’s - these are likely to go with the questions - “how do you do that, and why do you do it that way?”
- Competitor’s - digging into who they currently work with
Open vs closed
A quick moment to consider open vs closed questions.
An interview like this you want more open than closed questions. If you find you have it the other way around, try to think of how you could reframe the question to make it open.
For example:
Instead of “do you have this problem?” Opt for “tell me about the problems you have related to x”.
This will help you to reduce the risk of biasing the question. It then means if they mention the problem you want to help solve, you can ask follow up questions like “how big a problem is it for you?”
There will be times when you can’t avoid closed questions, but try to reserve them for follow-up questions.
Timing and sequencing
Ideally an interview with someone should take no more than an hour. I aim for 40 - 60 minutes. This is because you have to allow for lateness, housekeeping, and general chit-chat.
If you have a big pile of questions in front of you right now, you’re going to have to edit them because trust me when I say this, you won’t ask them all.
If you have a bunch of data gathering questions, you could ask them up front before the interview, either in advance or running through them really quickly at the start.
I would recommend you aim for 3 - 5 top level topic questions to cover. You can have a list of related follow-ups to them jotted down to help you, but this number is manageable for you.
You can have a list of bonus questions if you get what you’re looking for faster than you thought you might.
You need to remember that after the interviews you’re going to be analysing the data. Consistent questioning will make it much easier to compare how each person has answered.
Be mindful of the time you’re going to have, respectful of the time the participant is giving you and your own capacity.
Get those questions added to the sheet.
Hi there, have you been finding this guide helpful? why not subscribe to the Thin Slice so you can get the next in this series directly sent to your inbox.
3. Interview plan
Having an interview plan means you’re less likely to forget something, and also gives a clear structure to the session.
My interview plans look a little like this:
- 5 mins - intro, set the scene, housekeeping explain how I’m going to use this research. Ask if they mind me recording the session (so I don’t have to take notes and I have a transcript).
- 10 mins - question block 1
- 10 mins - question block 2
- 10 mins - question block 3
- 10 mins - question block 4
- 10 mins - contingency or bonus questions
- 5 mins - thanks and wrap-up
Feel free to steal this approach.
4. Consent forms
When you do research in a design studio, you get consent forms done. This covers your (delightful) derriere from the implications of GDPR if you’re recording the sessions. You need to give them the right to ask for the recording and any information you’ve gathered to be deleted.
You should also set out how you plan on using the research.
Generally speaking we shouldn’t use research sessions in marketing material. Your ideal client would be a little pissed off if they suddenly found themselves splashed all over social media.
If you’re creating a product, you also use it for confidentiality. If you’re doing this, PLEASE consult a legal professional.
5. Compensation
Are you going to compensate the person for giving up an hour of their time? Some people will gladly help you and not expect anything in return. However it’s respectful to compensate them for helping you out. Here are a few ways you could do that:
Money - good old hard cash opens a lot of doors. It’s up to you how much you want to offer.
Barter - trading an hour of your time / skills for an hour of theirs.
Freebie - if you already have a product you could offer them a freebie.
6. Recruitment
Recruitment is about understanding who you want to talk to, how many you want to talk to, how you find them, asking and then getting them to commit to it.
Who do you want to talk to?
You probably have a very clear picture of that person in your head. Which is great because I’d like you to jot that down (in the customer outline box on your worksheet)
It doesn’t need to be mega detailed, but it could look something like this:
Female business coaches in the UK who want to create a course
This is going to sound weird, but what you’re looking for it to do is help you reject people who don’t fit the brief. This one-liner helps you to do that. It tells you who, where and what.
If you want to bounce ideas of what this looks like or get help with it, remember to join the conversation so you can ask.
How many do you want to talk to?
I’ve got a boat load of experience with this, so trust me when I give you numbers to work with here.
Start with 5 people.
Let me break down the detail behind this number.
If you talk to 1 - 3 people you won’t learn very much. Hell, 1 person is a conversation with a mate, it’s anecdotal and doesn’t speak for a parish. Adding 2 more people to that mix sort of gives you a bit more information, but person 1 and 2 will probably tell you opposite things. How do you know which one to trust or which one is the consensus?
4 is OK, ish, but again, what if two people say one thing, and two people say the opposite?
5 people will give you patterns. Two could say one thing and three could say something different. Or all five could tell you something unique. If you talk to no more than 5 people you’ll have a ton of insight to help you move forward.
More than 5 people and you will find that they will start to strengthen the things people have already told you.
How do you find them?
Start with people you know. Reach out to them personally and explain what it is you’re doing (see why your plan was a good idea now?). You’ll find people who will be eager to help you.
If you’re struggling to find people who will help you that way. Try posting to one (or more) of your social media channels. Ask someone you know if they know anyone that can help.
Asking
What is it you’re asking them to do? You should have your plan, so asking them what you want of them should be easy. However, here are the things you should probably make clear.
- What do you want
- Why do you want it
- How long will it take
- When do you need it
- Are you going to compensate them.
Note if you’re posting on social media I wouldn’t share how much you’re willing to compensate them, simply because you’ll get people wanting the money, and you probably won’t get the insight you need.
Getting them to commit
Get it booked into both your calendars. Either you could send them a link to calendly or acuity (or whatever scheduling tool you might use) or crack open the calendars. Include the link to the session in there so it’s easy for them to join, include joining instructions if it’s an unfamiliar app.
You should also attach the consent form (if you’re using one) to the invitation.
7. Testing your research
Before you run the first research session, I would do a test run of it with someone you trust. This allows you to test out the flow of the questions. If you’ve missed anything. And if you’re happy with how it all hangs together.
This gives you the chance to tweak things before you run them for real. Better to iron out any glitches before you get going, than to find a problem after you’ve run one or two sessions.
Off you go…
I’m confident you have everything you need in order to carry out your first batch of research. If you’re stuck, got questions or need to bounce ideas, that’s what the chat is for, ask away!
Good luck!
In my next mail I’ll talk you through analysing your findings and how you can use them.
Advance warning (warning makes it sound terrible) I’m adding a new section to Unperfect Choices. I don’t want to use the term thought pieces, but as I’ve spent three hours writing this my brain is a little squiffy (sorry). I’m going to use it to talk about how I do what I do (my actual specialism rather than the generalist product stuff), how I identify problems and how I come up with ideas to solve them.
With that lovely thin slicers, I’m off to feed the Unperfect cheering squad (the cats and dog)
Emma
