Mine Customer Feedback for A/B Testing and Get 12.5% More Conversions
- 73% of website customers polled in a recent survey prefer getting live support via a chat box, as opposed to phone calls or emails.
- Exit intent popups are a great way of gathering A/B test hypotheses, especially since popups are 1,000% better for lead generation than sidebars.
- Products with user reviews have a 12.5% higher conversion rate than those without and are also 4 times more likely to be booked.
A/B testing and opinion-based User eXperience design have long been depicted as polar opposites. Some Internet marketing Singapore experts believe that you either scientifically test variants to find the best performer, or you just go with your gut feeling. You can’t do both.
Or, can you?
Back in 2011, as part of their April 1st prank, Unbounce crowned HiPPOS as rulers over the field of optimization. Not familiar with the concept? A HiPPO is the ‘Highest Paid Person’s Opinion’. It basically says that, testing or no testing, at the end of the day, it’s still your boss who calls the shots.
Image source: Unbounce
The joke was soon debunked and A/B Testing was reinstated as the most scientifically sound decision-taking process on website optimization.
Yet split testing for testing’s sake is not a good idea either: it simply provides quantitative data, with no actual explanation as to the ‘why’ behind the users’ actions. The (possible) solution? A marriage of opinion and A/B test results—but with the opinions collected via customer feedback.
Today’s post focuses on several methods for customer feedback opinion mining and also looks at one case study, in which feedback transparency actually boosted a company’s bottom line.
4 strategies for deriving A/B test hypotheses from user feedback
What split testing does: it takes some of the guesswork out of optimizing websites based on user intent and behaviour. What split testing doesn’t do: it doesn’t offer solutions on how to program real people like robots.
In other words, there’s no sure fire design strategy that can make users do exactly what you want them to do. In fact, the process works the other way around—your user experience design should be based on user behaviour.
The key is to know who uses your product, understand how they use it, and design accordingly.
That being said, knowing how to listen to your customers can be a difficult task. Luckily, there are several tools available to the internet marketing Singapore market, which make the process easier.
Live chat boxes
Users love live chat access, because they believe it’s the swiftest way to solve their customer support issues. This survey found that 73% of consumers prefer live chat over email (favoured by 63% of respondents) and telephone calls (44% preference).
More often than not, customers bring up small issues over live chat support. You don’t necessarily need to A/B test all of them—but, every so often, one such small detail could be tested to reveal major necessary changes for your website.
As their name indicates, these widgets are placed on websites to help gather user responses to your products, services, and site. They usually come in the form of a clickable box, placed in any one corner of the screen.
Image source: UserVoice
These widgets are usually highly customizable and can be set up to ask questions and opinions of your users. If you feel there are particular aspects of the site that are problematic and could use improvement, feedback widget questions are a great way to verify split test hypotheses.
Exit intent popups
It’s easy to set up an exit intent popup and based on the point in time at which they decide to leave your site, you can also ask them questions. The users’ answers to your questions will help you paint a more accurate picture of its strong and weak points.
Image source: Ben Philabaum
A note on popups: you would think most users hate them—but the surveys say they’re actually 1,000% more efficient for lead generation than sidebars.
Dropbox famously collected feedback on their product from user forums around the time they first launched it. They visited competitor forums and also asked HackerNews. In the end, the development and tweaking process they followed was more sophisticated, but forums were its basis.
You can also tap into sub-Reddits and ask redditors how they feel about your product/site. Moreover, forums are a great medium for spotting oddly isolated issues and gathering feedback on them.
Image source: Reddit for Brands
The impact of user reviews on sales volume: Figleaves.co.uk case study
Not only is user feedback great for collecting ideas for A/B tests, but it also goes a long way in promoting brand loyalty, conversions, and ultimately sales, too.
As the figures below show, figleaves.co.uk had a lot to gain from implementing a Ratings and Reviews solution on their website. Products with reviews are 4 times more likely to be booked and a 12.5% better conversion rate than those without.
Image source: Bazaar Voice
Products with more reviews (20 and over) have an 83.85% higher conversion rate, while the look-to-book ratio is 32.6% worse for products without reviews.