Winners & Losers: Split Testing Upsells w/ Ezra Firestone EPISODE 1 – BOOMSTICK TRIO

February 12, 2020 4:25 pm | Published by |

Winners & Losers: Split Testing Upsells w/ Ezra Firestone EPISODE 1 – BOOMSTICK TRIO

By Ezra Firestone | February 12, 2020

Hey, Ezra here from Zipify OneClickUpsell! 

This is the first post in a new blog series called Winners & Losers where I show you how I use split testing to optimize my Shopify store.

Specifically, I’ll show you the split tests I’m running to make more revenue from my most popular upsell funnels — 

Starting with my store’s flagship product, the Boomstick Trio.

You’ll see the winners and losers from 3 split tests I’m running, and then I’ll show you how I use that information to create follow-up tests to keep improving my results.

Optimization is a never-ending process, but my goal with this post is to give you as much insight as I can into how I optimize my $85 million Shopify store — 

So you can apply this knowledge to grow your own ecommerce brand. 

Quick Tip: If you want to run these same split tests on your store, you can use the Shopify app I created called Zipify OneClickUpsell. So far it’s made my store an extra $3 million in upsell revenue. 

Learn more about OneClickUpsell here.

The Boomstick Trio Upsell Funnel

For any given product a customer purchases, I’m using OneClickUpsell to offer at least one upsell offer and — if they say no to that — a downsell offer.

These post-purchase offers are what I’m referring to when I say “upsell funnel”.

This upsell funnel has three steps:

Step 1: The customer purchases a Boomstick Trio (yay!):

Step 2: I show them upsell #1 that offers a second Trio at a 21% discount:

Step 3 (Option 1 of 2):

  • If they say NO to upsell #1, then I show a downsell for my second most popular product at a 15% discount:

Step 3 (Option 2 of 2):

  • If they say YES to upsell #1, then I show upsell #2 for Boom Nectar at a 15% discount:

Then I send them to a Thank You Page, and the funnel is complete. 

(If I wanted to add more products to the funnel, OCU actually allows for up to 3 upsells and 1 downsell for every product, but I don’t always use that many.)

Now let’s start testing! First, I’ll walk you through the split test I’m running for upsell #1. After that, I’ll show you the new test I created to continue optimizing this offer.

Upsell #1: Single Product vs. Bundle

In this split test, I’ll test if a single product outperforms a two-product bundle with the same discount. The test looks like this:

Control Offer: Add one more Trio to your order now and receive 21% off. 

Challenger Offer: Add two more Trios to your order now and receive 21% off.

OneClickUpsell automatically sent 50% of the traffic to the control and the other 50% to the challenger page.

Quick Tip: Wondering why the product for upsell #1 is the same one they just purchased?

After generating $3 million in upsells on my store I’ve learned that no matter what you sell, the best upsell is (almost) always one more of the same product. You can read more about that here.

And the Winning Offer Was…

The control page!

The two-product bundle was only worth $3.91 per visit, whereas the control generated $5.32 per visit — a significant difference of $1.41.

(As you can see by the view count and total revenue of the control page, this variation has beaten a lot of challengers.)

So I’ll select the control page as the winner and create a new offer to test against it. 

Quick Tip: While the winning offer happens to also have the higher conversion rate (8.2% vs 3.25%), this won’t always be the case. Higher ticket items usually have a lower take rate but they can still result in more revenue for your store.

This is why I optimize for $/Visit because I want offers that generate more revenue, not necessarily more purchases.

Follow-up Test: Single Image vs. Carousel

Now that I know offering one Trio is better than two, I’m going to see if I can further improve the offer with a small design change. 

Control Offer: The same control page from the previous test.

Challenger Offer: Replace the single product image in the buy box with an image carousel of 10 photos. 

To do this, I’ll keep the offer the same for both pages (including the discount, copy, etc.) and I’ll just select image carousel inside of the app:

Once I modify the new offer, I’ll run this test and share the results in a later video.

Downsell #1: Fixed Size vs. Variable

The next split test in the Boomstick Trio funnel is for the downsell offer, Boomsilk. 

I’ll test if it’s better to offer a fixed size (4oz) of the Boomsilk moisturizer or to allow buyers to select their own size (2oz, 4oz, or 8oz).

Control Offer: Add a 4oz jar of Boomsilk and receive 15% off.

Challenger Offer: Select a 2oz, 4oz, or 8oz jar of Boomsilk and receive 15% off.

While a 4oz jar has a fixed price of $37.40, the challenger page lets the customer choose between three prices, ranging from $27.65 to $54.40.

And the Winner Was…

The control page was the winner once again, as it was worth 26¢ more per visit.

So even though the conversion rates were very close on both pages, a lot of visitors chose the less expensive 2oz jar which decreased the$/Visit to $1.01.

Follow-up Test: 4oz vs. 8oz

Now that I know offering a fixed size is more valuable, I’m going to change the size of the downsell and create a new challenger that tests if a larger and more expensive 8oz jar will outperform the control. 

Control Offer: The same control page from the previous test.

Challenger Offer: Add an 8oz jar of Boomsilk and receive 15% off.

This is super easy to do with OneClickUpsell; just select 8oz only from the list of variants:

And in this case, I’m also going to make a second change to the layout of this offer page and swap out the single image for a carousel.

Once I modify the new offer, I’ll run this test and share the results in a later video.

Long-form vs. Short-form Copy

So far, you’ve seen the split tests for upsell #1 and downsell #1 in my Boomstick Trio funnel. Now I’ll show you the final split test for upsell #2, Boom Nectar

In this split test, I want to know if an offer page with long-form sales copy outperforms an offer page with short-form copy:

Control Offer: An offer page with short-form sales copy for Boom Nectar.

Challenger Offer: A long-form offer page with an additional section of sales copy about the product.

Here is the sales copy I added to the challenger page (the control page can be seen above):

And the Winner Was…

In this case both variations were about even, with the long-form (the bottom variant below) slightly winning by 2¢.

So, basically, this test was a wash: Short-form vs long-form did not influence this particular offer. 

Hey, it happens. Now that I know the amount of sales copy I use doesn’t improve this offer, I’ll move on and change something else.

Quick Tip: There can be just as much value in learning what doesn’t affect your offer as what does.

Follow-up Test: 1oz vs 2oz

In this split test, I’m going to focus on a different variable than sales copy and test if a larger $73 bottle bottle of Nectar will outperform the 1oz version that sells for $49:

Control Offer: The same control page as the previous test.

Challenger Offer: Add a Boom Nectar 2oz to your order and receive 15% off.

Remember that, because the 2oz is a much higher price point, it can win this test even if its conversion rate is lower. 

Once I modify the new offer, I’ll run this test and share the results in a later video.


Want to Add Upsells to Your Store?

So far I’ve made an extra $3,138,214 in revenue for my store by adding post-purchase upsells like the ones I just showed you… 

And on any given day, OneClickUpsell adds about $50 to my average order value.

If you want to add post-purchase upsells to your Shopify store, visit http://zipify.com/apps/ocu to learn more.