As you probably know, the photo you choose for your Amazon listing can mean the difference between losing a customer vs making a sale. And more importantly, the quality of your photo determines whether you’ll even get a click on your product in the first place!
Speaking from experience, I’ve run experiments on my online store where a simple photo swap resulted in a 2X increase in conversions and an equivalent increase in Amazon sales rank.
But how do you choose the “best and highest converting” photos for your listings? How do you know that you’re using the most profitable images for your products?
Here’s a simple way to find out…
Get My Free Mini Course On How To Start A Successful Ecommerce Store
If you are interested in starting an ecommerce business, I put together a comprehensive package of resources that will help you launch your own online store from complete scratch. Be sure to grab it before you leave!
The Challenges Of Testing Your Photos
When it comes to finding out what converts the best, you “usually” have to perform some sort of split test.
And for all of you who are not familiar with the concept, split testing is where you take 2 different samples, show them equally to your customers and figure out which one performs the best. In fact prior to this year, it was the main way I gathered data for my ecommerce store.
But the problem with traditional split testing is that it takes forever.
Every test that I run usually takes at least 3 weeks or more and I’d say that 9 times out of 10, my tests are inconclusive.
The other problem with split testing is it requires you to insert a conversion pixel or some sort of tracking mechanism on your site so that you can measure an actual sale.
Anyway, the main issue with this requirement is that sites like Amazon.com do not let you do this at all. As a result, it’s extremely difficult to run a good split test on your Amazon listings.
Testing Your Amazon Listings
Have you ever wanted to test different product images on Amazon to see which one converts the best?
Have you ever wanted to test different bullet points to see which is more appealing?
If you are an author, have you ever wanted to test different book covers to see which one is more clickable?
Right now if you were to do this, you would have to put up a listing, wait a few weeks, swap out the images for new ones, wait another few weeks and then compare the results.
Not only is this a major pain in the butt if you have multiple listings, but if you’re lazy like me, you’re never going to do it.
In addition, Amazon doesn’t let you put any special code on their site so traditional split testing methods are impossible.
So this got me thinking…
What if there was a way to solicit real user feedback about a product listing BEFORE even putting it up on Amazon?
Sure, the results wouldn’t be as accurate as a true split test. But most likely it would be good enough.
How One Podcast Guest Changed Everything
A while back, I interviewed my buddy John Li on my podcast where we discussed how he created the popular tool PickFu.com.
And in case you don’t remember what his tool does, it basically allows you solicit real human feedback about anything within 10 minutes. For example, if you have 2 headlines and you want to know which one is more appealing, you can get feedback immediately.
If you have 2 website designs and you want to know which one is more attractive, PickFu will get hundreds of people to provide their opinion in under an hour.
If you want to know whether a product you want to sell has any potential demand, you can get an unbiased opinion.
Anyway after our interview, we started talking about potential applications in the ecommerce space and a huge light bulb went off in my head.
Why not use PickFu to get feedback on an Amazon product listing BEFORE it goes live?
Soliciting User Feedback
So I thought I’d give it a try just for fun.
Below is a product that I currently have listed on Amazon and I decided to use PickFu to see if the current product photo could be improved.
Here are the 2 photos that I decided to compare
Setting Up The Test
Just for the record, I normally hate setting up and running traditional split tests because they are a major hassle.
But I was pleasantly surprised by how easy it was to setup a PickFu test. And the results were incredibly fast.
There are 3 basic steps.
- Enter a poll question
- Upload 2 things to test
- Select your demographic
- Hit go
- Wait 20 minutes and get instant results
And the best part is that it’s super cheap to run a test. This particular poll cost me $16.
By the way, John was kind enough to extend a 50% coupon code for all MyWifeQuitHerJob.com readers.
Please use coupon code: MYWIFEQUIT to get 50% off (Click here to run your first poll)
Note: Normally this poll would cost $25 with the coupon. But I chose to target only females for my poll. There are a number of ways to restrict your test within a certain demographic but it costs more.
The Results Of My Poll
Here are the results! I started this poll at 8:48am on a Sunday morning and got results at exactly 9:09am. That’s like 20 minutes!
Keep in mind that the image on the left is the ACTUAL image I’m currently using on Amazon. But after running this poll, I’m definitely going to switch the photo in my listing.
As you can see above, the second picture won by a landslide.
But it gets even better.
Every single polled user also leaves verbal feedback as well! So not only do you get the results but you also get an unbiased explanation for the vote count.
And finally, you also receive demographic information on who took the poll. If I were to do this all over again, I would probably restrict my poll to a wealthier audience because those are the customers I tend to attract in my wedding linens store.
Traditional Split Testing Vs PickFu
Are the test results from my PickFu experiment definitive? Probably not.
But if a group of unbiased people are 3.16 times more likely to choose image B, then there’s a very high likelihood that image B will perform better on Amazon.
In any case, I want to say that a tool like PickFu will never ever replace traditional split testing. After all in a real split test, you are testing against real customers on an actual website.
But here’s the thing.
I’ve run a bunch of split tests in my day and they are major pain in the butt to setup even with the right tools.
Just recently, it took me several precious hours away from my family to setup a photo split test for my ecommerce store which is still in the process of running.
Meanwhile, split tests also take a long time to get results. And depending on your traffic levels, it can often take several months before you can get an answer.
Now it’s one thing if they always ended up providing conclusive results, but it’s rarely the case. And I would go as far as to say that over 90% of my tests don’t tell me jack.
So from my perspective, I’m very reluctant to run a split test now unless it’s for something major. But for everything else, I’m more inclined to pay 10 bucks and get a quick 20 minute answer from PickFu.
Seriously, for something like Amazon where split testing is impossible, it’s worth it to pay 10 bucks and get an immediate opinion on your photo or listing because there’s a lot of money on the line.
Especially for sites that are not conducive to split testing at all like Amazon, EBay or Etsy, you should definitely solicit some unbiased feedback about your listing before it goes live.
At a bare minimum, you should test your product images because they are arguably the most important part of your listing. And if you plan on making a living selling on Amazon, it also makes sense to test your bullet points as well.
If you run your own website but get very little traffic, split testing is often not an option either. For example if your site gets less than a few thousand hits per month, it might take you over a year to gather enough data to run a test and 90% of your experiments will yield inconclusive results.
By soliciting some quick feedback with PickFu, you can easily address 80% of your conversion issues in a matter of minutes.
So if you sell on Amazon, EBay or Etsy or if you run a low traffic website, I’d give it a shot and check your results.
BTW, if any of you have any split testing stories to tell, please enter them in the comments below. I’ve got a lot to share in this department because split testing, while useful when it actually finds something, has been frustrating for me in the past.
It will only cost you 10 bucks and the results could improve your Amazon sales dramatically.