What happens if you show a coupon code in the shopping cart?
Recently I performed a very interesting A/B test, involving a variation showing a coupon code on the shopping cart page. The hypothesis was that the coupon code section could drive away customers that don't have such a code. So what happens if there is a coupon code on the shopping cart that can be directly used? Here you see the variation:
In this case, an additional €10 discount is offered when this code is applied. We saw a big increase in conversion (I can't tell the exact numbers, sorry). Because the profit per conversion is lower in the variation, it's very interesting to see if total profitability is increasing.
For example, suppose the average margin on each order is €100 without any discount code. In the coupon code situation this is € 90. In this case, increase in conversion rate should be at least 10% to justify the decision. But what if the average margin for each order is €30 without the discount (and €20 with the discount)? In this case, an increase in conversion of at least 33,33% is needed. Something to keep in mind...
Do you have any experience with testing coupon codes like this? And what were your findings? I'm very curious....
I have often thought about the effects of a discount box in the checkout and the minefield that it offers.
One thing I have always wanted to do is random discounts based on the value/profit of the visitors cart. So if they buy it right there and then they will get X discount applied to their order. This can be based off an algorithm that predicts the level of discount to offer to seal the purchase based on business factors and past performance.
I would be interested to hear what testing people have done in this area.
So, I have ran a similar test. My test actually has a picece of content in the cart as a friendly reminder to "use coupon code for free shipping on orders over $75" . (see screenshot).
Your discussion actually made me turn this back on to get more results.
Below are my current results thus far:
- Engagement is down by 0.3%
- Revenue has increased by 1.4%
- Clicked Proceed to checkout has increased by 0.2%
Granted, these are not huge numbers and I have not reach statistical significance. I'll keep you guys posted on this.
My next iteration will be to apply the discount automatically when the user gets to the $75 free shipping threshhold.
What is your current average order value? Is $75 the right price-point to make this most effective?
Do you offer easy cross-sell and up-sell additions in the checkout to help people reach $75?
Great to see you're also experimenting with this. The result you see, have you segmented the data to spot differences on traffic source/visitor type/browser type or any other segment? Maybe you can spot some interesting differences and is something having a negative effect in this variation (which explains the small gains).
And maybe you can experiment with some urgency triggers like: "order today and ship it for free: use code "FREESHIPPING"? Or maybe limit the amount of people that can use the code ("First 15 orders today get free shipping").
And as a next step you can do a sidewide test with this message to see if this also generates extra sales?
I've done some testing around checkout promo codes, but using a slightly different experience. Rather than by default presenting a code to use, I've taken the approach of obfuscating the input behind an accordian/toggle, and only having it on the checkout page. The intent is that the user sees 1 less field (increase form conversion) and to de-emphasize the promo code input so that it's not right there in the users face causing them to wonder "am I missing out on a good discount?"
Experiments run on 3 distinct e-commerce sites so far, each with drastically different products and purchasing habits. Performed both during average traffic periods and during high holiday periods. Results so far have been inconclusive, but at worst it has had no negative impact on revenue or order conversion. In each case, instances of orders placed with a promo code have gone down for the variation that obfuscates, so theoretically less margin is being given up. This is the kind of exeriment where you really need outside analytics/data beyond what the test itself can provide.
So the people who have come to your checkout and already have coupon codes get a slightly different version to those who do not have a code.
I am not sure how you would separate those visitors? Maybe by asking 'Do you have a voucher code?' then by the 'No' answer have a 'click here to get one' section. You can then show visitors a selection of offers to pull them over the line.
This would help to prevent them heading off-site hunting for a discount!
Our AOV is around $72.68 right now. Based on this comment I actually ran a free shipping test for orders over $99. It didn't do well at all in terms of Revenue.
I'm scratching my head on this one.