The conversion rate optimization niche on the web is full to bursting with advice on what you should change on your website to increase your conversion rate.
Whilst it can help spark ideas when you see what worked for other websites you shouldn’t listen too hard.
Every second you spend reading about someone else is time you’re not spending improving your own site.
All websites are inherently different. They have different reasons for being, with different customer groups, in different markets, styled in infinitely different ways and their objectives are nearly always completely different.
And you certainly shouldn’t blindly copy.
A cursory glance of a listen should be just enough.
Talk to your customers
You need to be led by your customers past, present and future. They know you best.
Do you want to know what they think?
You should ask.
Asking them the right questions will tell you all you need to know about how you should be marketing yourself and also what products or features you should be working on next.
But … you need to be open-minded and humble enough to listen and benefit.
Up for that?
This is what CRO geeks mean when they say they’re going to uncover your ‘barriers to conversion’ and remove them one-by-one.
This is the first post in a series
I’m going to take detailed looks at ways to talk to your customers to literally cover you in customer voice data for your next conversation rate optimization push.
What a joy.
Why did your customers buy from you?
Sending a survey to past customers to hear why they loved you enough to buy from you is one of the easiest ways to start building a mountain of customer voice data.
It’s also very easy to mess up.
One false move now and you run the risk of annoying ALL your past customers.
Before you send anything to anyone take the WIFT test (What’s In It for Them).
Rule #1 – Only send the survey from YOU
When other departments hear that you’re about to send a survey out to customers they’re going to want to get involved. There’ll always be someone somewhere wanting to sneak an extra question in.
But. Like your email marketing. Every survey should only ever have 1 objective and if yours is to increase your conversion rate then stay laser focused on that and block out the noise.
How do you do that?
Get political cover first before you announce a survey is going to go out.
Go right to the top of your company. Explain your strategy and why it’s important that the questions you are going to be sending should be the only ones.
Explain that you’ll be sending only open free text questions so there will likely be lots of intel for all the other teams to feast on anyway.
Once you’re covered you’re good to go.
First thing you need to do is choose a survey tool
Don’t be tempted to do this yourself from your own CRM system or via your email marketing channel. To ensure you get best results you need to be using the right tool for the job. Variety of hammer for nail so-to-speak.
The great thing about this startup-rich world that we live in is that there are plenty of decent options available.
The big chief in town is Survey Monkey
Survey Monkey is a solid choice. They’ve been around a long time so they really know their stuff but if I’m honest I haven’t had the best experience with them.
The functionality is great and their scale now means their price point is excellent but because they’re a big company their support doesn’t seem to respond as quickly as it used to.
I hate losing time over trivial holes in the road and having to wait for my support ticket to make it to the bottom of the funnel is a big personal peeve.
If you’re like me and you want everything to happen immediately then there’s a good alternative.
I like Fluid Surveys a lot
Fluid Surveys have a stunning user interface and their survey builder tool is actually quite a lot of fun to use as well as being feature rich without being overbearing.
The first time I used them it took 10 mins to put the foundation of the survey in place. I then had to spend a whole 20 mins more to replicate it in 4 other languages for my client.
Once you’ve created your survey you can style it, upload your customer list into their system and send out emails with unique survey links that make it super easy for your targets to fulfil their end of the bargain.
Then, as the results come in, they have a tidy little reporting interface where you can view answers, generate word clouds, download spreadsheets and all sorts of other good stuff.
The only thing I don’t like about them is that they have a limit on the number of emails you can send from their system. This made no sense to me when I came across it. Why would you throttle down on the only way to get your customers to use their tool?
It’s because they’d had some problems with spammers (there’s always someone trying to ruin the party).
You can have this limit removed if you talk to them about it.
Guidelines for content and survey structure
I have a few general guidelines for the content and structure of a survey:
- Make your survey only 1 page (like with a checkout process every additional page is a conversion killer so don’t let that happen)
- Have no more than 7/8 questions in total – it should be possible for a customer to complete this survey in under 5 mins if they really want to
- Most questions should be of the open-ended free-text variety rather radio buttons, dropdowns and multiple choice (you want to hear their words rather than force them to use your voice)
- The questions should invoke their imagination and be fun to complete rather than dry, boring and predictable
- Only include 1 quantitative question to help you measure progress over time (Net Promotor Score is the one I use)
What questions should you be asking?
After a few years of doing this my favorite set of questions are these:
- How did you first find out about us?
- What persuaded you to buy Product X from us?
- How would you describe Company Name to a friend?
- If you could change one thing about our website, what would it be?
- What would persuade you to use us more often?
- If you could have us create a product or service just for you, what would it be?
- How likely is it that you would recommend Company Name to a friend or colleague? (0-10) where 0 is not very likely and 10 is very likely
There are some hidden objectives and clever psychology behind the survey questions
1) How did you first find out about us?
This gets the responder warmed up. It’s an easy first hurdle for them to leap and the content will likely be short, to the point and most importantly tell you where your marketing is most effective.
If no-one is saying Google then you could reduce your PPC spend and focus on making your brand more easily passed by word-of-mouth.
2) What persuaded you to buy Product X from us?
This invokes a mild emotional reaction. We could’ve simply worded the question ‘why did you buy Product X from us?’ but to ask what persuaded them is a level up and gets them to think deeper.
Psych: To reveal what persuades us is far more emotionally challenging.
3) How would you describe company name to a friend?
This focuses the responder on the present. It could’ve again been more simple but it’s deliberately not.
We could’ve said ‘How would you describe us?’ but bringing their friends into the equation means they’ll think more carefully about the answer they give.
Psych: No-one wants to look silly in front of their friends and even though they’re not really involved in this little tete-a-tete the trick works well.
4) If you could change one thing about our website, what would it be?
This puts the responder in a position of power which they like (who doesn’t). Note here that we’re not asking for a multitude of things to change because that’d be a daunting.
We just want one. Everyone has at least one good idea, don’t they?
Psych: The emotional trick here is that we’re, in effect, putting them in charge of the company’s direction for a split second. People like power and influence.
If we wanted to be even more explicit about this we could ask ‘If you were CEO of ‘Company Name’ what would be the one thing about our website that you’d change?’
5) What would persuade you to use us more often?
This may even lead to a conversion of its own. It’s also bringing that powerful ‘persuade’ word back in again except this time it’s more forward thinking.
Psych: It’s putting them into a position where they’re thinking about buying from you again and forcing them to think deeply about your company’s value proposition.
6) If you could have us create a product or service just for you, what would it be?
Psych: This is truly putting them in the power seat of your company and gets them to design your product line for you.
Your product team is going to be your BFF once they’ve seen the answers to this.
You probably won’t be using any of these responses to impact your immediate conversion rate effort but you should, at some point, be looking to change up your product mix and what better way to do that than by adding products that your already loyal customer base has heavily requested from you.
For example, if you get lots of ‘I’d like free shipping’ then you need to find a way to give that to them.
7) How likely is it that you would recommend company name to a friend or colleague?
Number 7 is a nice easy wrap up question and our only quantitative question.
They’ve been thinking deeply about your company from a number of different angles over the previous 6 questions so they’re in a great position to now decide how much they love you or not.
This is the Net Promotor Score that is widely used across industry so you can benchmark yourself and measure this over time to make sure your product and service is improving.
For a good description of how the NPS works have a read of this.
Here are a few notable names and their NPSs. Where do you fit in?
- Amazon 76%
- Apple 71%
- Costco 71%
- Virgin America 66%
- Google 56%
- Hilton 55%
- American Express 43%
- Trip Advisor 33%
- Facebook 31%
- American Airlines -5%
- Mediacom -21%
The order of these questions matters
You need to make the story flow and feel right. Even for a survey.
The reason the survey questions are in this order is because they lead the responder along a chronological path. Initially they think about the past (Q1 & Q2), then the present (Q3) and finally the future states (Q4, Q5, Q6 & Q7).
Make sure you provide an incentive
Here’s the WIFT. Everyone needs a little encouragement to complete a survey. There are very few companies that people love so much they’re happy to do them for free.
But that doesn’t mean you can’t have a little fun.
Forget the iPad. That’s so last decade. In fact anything Apple related has been done to death.
How about using your own product as an incentive? These are your customers after all so they liked what you have to offer enough at least once.
If you sell big ticket items then you could offer their last purchase refunded.
If your prices are a little cheaper then how about a nice chunky voucher.
Setup a unique-for-purpose ‘from’ email and monitor it
You need to send this survey from an email address that has been setup for this activity only. Don’t send it from yours, or your CEOs or even your worst frenemy’s email because there are going to be lots of questions back.
Is this spam?
I don’t quite understand the point of question 3.
Is it OK if I only complete 5 out of the 7?
Have I won?
They’ll come flooding in and you want to be answering them because they’re your customers and you want to keep them onside (or get them onside if they’ve previously had a bad experience with your company).
Your survey invite email should be short and clear
The format I tend to used is one I borrowed from the Conversion Rate Experts and their sunshine.co.uk case study.
It’s easy to understand, personal, emphasizes benefits and gets a great response rate so no need to re-invent the wheel.
Make a big deal of the prize presentation
It’s very important to close the feedback loop so make sure you get some mileage out of awarding the prize.
Present it in person to the winner if you can. Make it a surprise and video the whole thing. At the very least take some photos and blog about it after.
Then update your whole list with this material so they can all see that you follow through on your promises.
Because when you do it all again next year you’ll want to have another strong response rate.
Launching a successful survey is an art that shouldn’t be left to chance
Follow the above steps and you’ll get a good response rate, some excellent ideas, a tonne of social proof, and maybe even a few additional sales along for the ride.
Have I missed anything?