Read on Twitter
So you know you need qualitative customer research - or maybe your client does. Great!
Here’s how to start your customer research off right, a speedy guide (thread
)
Here’s how to start your customer research off right, a speedy guide (thread
)
Warning: What you're about to read is a step by step of what to do BEFORE you start any kind of qualitative research.
This is is guaranteed to be the least sexy - and most important - part of the research process.
This is is guaranteed to be the least sexy - and most important - part of the research process.
Do this well and you’ll get accurate, reliable customer insights in surprisingly good time.
Skip it or rush through and you’ll end up with insights you can’t use, can’t trust or that don’t ultimately lead to increased conversions.
Ok let's start!
Skip it or rush through and you’ll end up with insights you can’t use, can’t trust or that don’t ultimately lead to increased conversions.
Ok let's start!
All good research starts with a good research question.
In case this seems like unnecessary formality, it isn't.
You NEED to define your research before you start doing research.
In case this seems like unnecessary formality, it isn't.
You NEED to define your research before you start doing research.
Research without a research question is a crapshoot.
It's like trying to drive east without a route or a map. You can do it, sure, but it'll take you twice as long to get there - if you even get there at all.
It's like trying to drive east without a route or a map. You can do it, sure, but it'll take you twice as long to get there - if you even get there at all.
So - research questions. How do you come up with one? What’s a good research question?
A good research question does 2 things:
1. It spells out what you want to find out and
2. It only asks what’s in your means to find out.
1. It spells out what you want to find out and
2. It only asks what’s in your means to find out.
If the answer to the research question is out of bounds, if it's not possible for you to answer that question, it’s a bad question.
A good question is specific and actionable but complex (i.e., not a yes / no answer).
A good question is specific and actionable but complex (i.e., not a yes / no answer).
Bad research question: “Why do customers buy?”
Good research question: “What personal, circumstantial or product-specific factors are driving sales in the past 6 months?”
The bad one is vague with no direction. The good one is specific, tangible & sthg we can actually find out.
Good research question: “What personal, circumstantial or product-specific factors are driving sales in the past 6 months?”
The bad one is vague with no direction. The good one is specific, tangible & sthg we can actually find out.
Once you have your research question, think through how to answer it.
What evidence or clues do you need to find an answer?
Do customer interviews make sense? What about customer surveys? What point in the customer journey do you need to hone in on?
What evidence or clues do you need to find an answer?
Do customer interviews make sense? What about customer surveys? What point in the customer journey do you need to hone in on?
Let's take the research question from a few tweets back -
What methods can we use to find out what's driving sales in the past 6 months?
What methods can we use to find out what's driving sales in the past 6 months?
Whenever you're trying to understand a decision point try to get data from both sides of the decision i.e., pre and post purchase.
Then you can compare what you find pre and post to see if it lines up.
Then you can compare what you find pre and post to see if it lines up.
Here are the most common qualitative customer research sources / methods:
- customer interviews
- surveys
- on site polls
- mining product reviews
- sales / demo / discovery call analysis
- live chat transcript analysis
- customer service ticket analysis
- customer interviews
- surveys
- on site polls
- mining product reviews
- sales / demo / discovery call analysis
- live chat transcript analysis
- customer service ticket analysis
**Customer interviews are all the rage & def insightful, but not always possible or a fit for what you need. For ex, with customer interviews you're talking to a much smaller sample (think 7-10 ppl) which isn't always or necessarily representative of your entire customer base.**
Another side note:
We're talking primarily about qualitative research here - not quantitative, which is more abt Google analytics and heatmaps.
Those have an important place in research for sure but are not the focus of this thread.
We're talking primarily about qualitative research here - not quantitative, which is more abt Google analytics and heatmaps.
Those have an important place in research for sure but are not the focus of this thread.
Let's go down the list of methods.
- Product reviews: They're more about life post-purchase & not about the decision to buy
- Customer support tickets: Could be good to see common Qs ppl have before they buy.
- Live chat transcripts: Same as customer support tickets
- Product reviews: They're more about life post-purchase & not about the decision to buy
- Customer support tickets: Could be good to see common Qs ppl have before they buy.
- Live chat transcripts: Same as customer support tickets
- Surveys and interviews: Could work; Surveys to give us reach & a big sample size, so data is representative. Then we can hone in on what we learn from surveys with interviews.
- Talking to sales & customer service agents: Good to validate themes
- Talking to sales & customer service agents: Good to validate themes
- Sales call recordings: Could help validate themes but will probably be redundant - unless we're not finding any patterns in the other methods
Ok! Now we've landed on:
- customer surveys & interviews to start
- then sales & customer service rep interviews
- a look at tickets & chats too, to supplement what we learn from the customer-facing employee interviews
- customer surveys & interviews to start
- then sales & customer service rep interviews
- a look at tickets & chats too, to supplement what we learn from the customer-facing employee interviews
And there you have it! A robust & strategic customer research plan with clear rationale you can share with a client.
You don't have to do/use EVERY method but you have a few to choose from and a priority list you can use or share with a client.
You don't have to do/use EVERY method but you have a few to choose from and a priority list you can use or share with a client.
Research can sound pretty vague & elusive.
If you're doing research for a client, the more transparency you can give into your research thinking and process, the more confidence they'll have in it being a worthwhile pursuit, and in your ability to generate insights.
If you're doing research for a client, the more transparency you can give into your research thinking and process, the more confidence they'll have in it being a worthwhile pursuit, and in your ability to generate insights.
Once you go through this process a few times it'll be pretty fluid, and you can spend more time on honing the exact questions to ask in each interview or survey - which is where the juicy work really bigs!! 
I hope this was helpful! Questions?? Thoughts??
Drop em below
I hope this was helpful! Questions?? Thoughts??
Drop em below
