Cross selling and up-selling

If your site offers multiple products, add-ons and extra services, then cross selling and up-selling are important ways to maximize your profits.

At, we sell seven software products (each available in two editions), five iPhone apps, two different barcode scanners, plus we offer CD delivery and a paid support option called “Priority Support”.

The software products are all related and targeted at the same type of customers, so cross-selling has always been important for us. And with the introduction of the barcode scanners we added some great stuff to up-sell to new and existing customers.

Before telling your about our current solution, let’s look at the various methods for cross selling and up-selling we tried since we added our second program.

Overall Buy page with extra cross-sell interstitial

In an attempt to make every buyer aware of all our software products, we always took them through our overall Buy page, a product-selection page showing all our programs. It had a prominent message about our bundle discount (20% if I remember correctly) pushing them to select multiple products.
Stubborn customers who still dared to select only one program, got hit with an extra cross-sell interstitial page, reminding them that they could add a second product for only x dollars.

This worked great for a while. But our Buy page got more and more complex, because we added more programs, added the barcode scanners and started offering two editions (Standard and Pro) for each of them. Which didn’t exactly help to improve our conversion rate.

Pre-fab 3-program bundles, straight to Checkout

Then, by accident really, we found a better solution.
We wanted to do a special Christmas Offer, so we created a bundle of our three most popular programs (Music Collector, Movie Collector and Book Collector), the laser barcode scanner, a software CD, plus one year of Priority Support. All for just $199.95, at the time that was a $60 discount.

The special bundle was prominently listed on all product home pages, its Buy button taking buyers directly to the Checkout page, skipping both our complex Buy page and the cross-sell interstitial page.

The offer was a huge success, so on December 27, we just renamed the offer to New Year’s Offer, extended its end date to January 31. And it kept selling, like crazy.
This “temporary” offer has been up on our website for several years, we just renamed and extended it every couple of weeks. We celebrated everything, Spring, Summer, Easter, Valentine’s Day, Mother’s Day, Father’s Day, new software versions, big movie releases, etc.. I’ve seen some pretty creative offer names 🙂

Single program scanner bundles, straight to Checkout

Last year, Sytske and I were at a Perry Marshall seminar. After out “hot seat”, presenting the Movie Collector product page to the group, our friend Joshua Boswell made the following observation:
The prominent presence of the Special Offer bundle could be diluting our DVD cataloging message. By immediately confronting movie fans with the possibility to catalog their CDs and books too, we may be confusing them, making them check the other software first.

Joshua suggested to have the product page focus more on the specific product or solution that people had been looking for.
“They came looking for a DVD cataloging solution, so sell em Movie Collector. Up-selling to barcode scanners, great. But no immediate cross-selling. Do the cross-selling on the back-end.”
(Both Joshua and Perry are big fans of selling stuff on the back-end using auto responder email sequences. Nothing is easier than selling to existing customers.)

So we did. We stopped doing the Special Offer and replaced it with two new pre-fab bundles, both with Buy Now buttons that took the buyer directly to the checkout.
A low-end bundle with Movie Collector Pro, a cheap barcode scanner, CD delivery and Priority Support, sold for $89.90. Plus a similar package, bundling Movie Collector Pro with the expensive laser barcode scanner, going for the total of $219.95.

Yes, that’s right. That’s $20 more than the old Special Offer, but having only one program instead of three. Okay, by then, we had replaced our laser scanner with a better and slightly more expensive one, but still… No discounts whatsoever, everything full price.

And amazingly, it worked.
That is, I was amazed, Joshua was more like “I told you so” 🙂
The changes increased both our conversion rate and our average first purchase value (the front-end sales). Of course we sold less software licenses, but that was great because it left us more stuff to cross-sell on the back-end.

Simpler bundles, via up-sell interstitial page

However, in the past few months, two things began bothering me, issues that I suspected were hurting conversions.

First, the lack of flexibility: some users simply don’t want or need the CD or the support service. Or some do want to bundle the scanner with two programs instead of one. Of course, that was still possible by going through our complex Buy page, but not a lot of users found it (or understood it).

Also, I wasn’t happy with the $219.95 price tag of the laser scanner bundle. While we were doing the 3-program Special Offer, we had experimented with its price and setting the price just below $200 sold best.
But I didn’t want to just lower the price of the bundle, cutting my margin with $20. The only way to get it to $199.95 would be to remove the the CD-delivery and priority support from the pre-fab package.

So here’s what we did: we stripped all prefab-packages down to the basics, either just the program, the program plus a cheap scanner or the program plus a deluxe scanner. Then I made their Buy buttons go to a new interstitial page that offered relevant upsells only (as opposed to the Buy page that just offered everything else).

Of course, the obvious upsells to present there are the CD delivery and the priority support. For software only purchases, both scanners are great entries under “Other Movie Collector customers also bought:”.
And for orders that already include a scanner, I now sneakily try to cross-sell to our other scanner-enabled software, figuring “hey, if you are going to buy a cool piece of hardware, why not get full use out of it?”.

Here’s a screenshot of the upsell page, showing the offered extras based on a software-only order:


Or try it live here. Tip: try adding either one of the scanners.

Now, you know I don’t make drastic website changes like this without A/B testing, especially not for pricing and checkout related changes. You never know what might happen. In this case, I thought there were the following risks and potential benefits:

– Less CD and Support sales because these are not included by default anymore. Possible resulting in lower first purchase values.
– Lower conversion rates because of the extra step in the checkout process.

Possible benefits:
– Higher conversion rates because of lower prices of the pre-fab packages.
– More scanner sales because they are more aggresively pushed to people initially opting for just the software.
– More software sales because of the cross-selling to scanner buyers.

So in the past two weeks, I have been testing the old situation (= A) versus the new setup (= B). For clarity:
A: Program + scanner bundles, including CD and Support, clicking straight to Checkout.
B: Basic (and cheaper) program + scanner bundles, taken through the up-sell interstitial page.

Result for Up-sell Interstitial A/B test

This was quite an important test, so to be sure about the results, I let it run for two weeks. During these 14 days I tracked sales for a total of 506 new customers (customers who visited the website for the first time within this period).

Amazingly, the distribution over A and B was 253 vs 253, meaning that the conversion rate was more or less the same in both groups (not exactly the same, because the number of visitors in B was slightly lower, making B’s conversion rate 0.67% higher).

Disappointing? Not really, because the Average First Purchase Value in group B was 10.3% higher, making the total profits in group B 11.4% higher than those in group A.

Other results: The number of people who opted for CD delivery stayed the same, but Priority support purchases are down somewhat. Also, we sold exactly one Mouse pad 🙁


Don’t hesitate to push people to buy more stuff during checkout, as it can help to squeeze out a couple more dollars. But be careful trying to cross-sell your other programs. Don’t offer them too prominently on your landing pages. Only offer them as up-sells during checkout when they are targeted. Otherwise, leave the cross selling until later and make them aware of your other software by email.

Furthermore, an extra step in the checkout process doesn’t seem to hurt the conversion rate. That is, as long as it offers sensible, targeted up-sells (no mouse pads!) and can be easily skipped. Or was it the lower bundle prices that compensated the loss in conversions caused by the longer checkout?
Mmm… I think I need to do more testing.

2 thoughts on “Cross selling and up-selling

  1. Hi Alwin,

    I finished my AB-testing with a shopping-page very similar to yours. But the results weren’t so good. Depending on the test run (I tried 2 variations), the new version made either “slightly less” sales or “50%” less (which I quickly stopped of course). Strangly enough the worse version only differed by the added money-back-guarantee image.

    Perhaps another case of “too many options”. So it didn’t work for me, but it still was interesting and I’ll be trying other things instead. 😉


  2. Pingback: New shop page «

Leave a Reply

Your email address will not be published. Required fields are marked *