In the previous experiment, we have seen that Trustpilot reviews didn't help to increase conversions. This was a surprising result.
In my career, I have never seen anything like that. So I had a phone call with our Trustpilot Account Manager and we discussed the reasons for this failed experiment.
What I have learned from this discussion were three things that triggered a second experiment.
This time the second experiment was a huge success.
Here is why the first experiment failed:
Our number of Trustpilot reviews were too small for a company of this size with thousands of students. Back in November 2019 we only had 323 reviews.
The average TrustScore was 4.5/ 5 which apparently negatively affected our reputation online
Our Trustpilot widget used for the experiment wasn't really allowing prospective customers to actually READ any review. It was just showing the total amount of reviews and TrustScore.
The new Trustpilot experiment setup
That said, the Trustpilot representative and I came up with an idea for a second test, where all these objections were resolved.
In April 2020, we had more than 439 reviews, a total TrustScore of 4.8/5 and we decided to use a different widget to test a different hypothesis.
Which widget could we use to accomplish this?
The Carousel review. It was the perfect candidate for this new experiment:
It displays your latest 15 reviews in a carousel
It is a non-clickable TrustBox ideal for reassuring the decision to buy
It has a transparent design that fits any background
The courses' landing page
Here is what it looks like:
New experiment hypothesis
The questions we wanted to answer with this new experiment were the following
Can the carousel widget displaying reviews increase our conversion rate?
If so, by how much?
The new results were more encouraging than the previous experiment.
The new experiment widget:
The main difference was that the widget was shown on the product landing page, so the result was a 21.59% conversion rate.
Without the widget, the new experiment only saw a 13.68% conversion rate, a difference of 44%.
We also tested two options with the carousel: the Carousel above, and the Carousel below the main products. We wanted to assess the impact of the carousel based on its position on the page.
The Carousel above proved to convert better, simply because it was placed above the fold, where 100% of people were seeing it.
This is further proof that the increase in conversions was due to the reviews being in a visible place.
More interesting findings we made were site usage stats, which changed massively.
Page per sessions: Users landing on the page without a Carousel visited on average 4.73 pages, whereas users landing on the page with a Carousel visited fewer pages. This can be explained by the fact that people reading reviews trust the brand more than those who don’t get a chance to read the reviews. People have read the reviews need less time to be convinced and to convert.
The average session duration: the pages showing carousel has a lower session duration, which confirms the above statement. Reviews increase trust in brands.
The bounce rate was higher on the pages displaying the carousel.
It's easier to draw some conclusions with this outcome. Reviews from real students, that people can actually read, are more likely to convey trust on the product.
It's very important that reviews must refer to the product, not to the company in general, to be more credible. People tend to care more about the quality of products and services delivered because those are the elements that eventually create the company's reputation. If your products are of bad quality, it probably means your company is too.
The higher TrustScore of 4.8/5, if compared to 4.5/5 in November 2019, made a huge difference in perception.
Online reviews are super important as a conversion rate factor, but only if displayed in a way that offers added value during the decision-making process of each individual.
These tests seem to suggest that the right type of online reviews can be instrumental in driving more conversions, at least according to an online courses provider.