Make money from your checkout completion page!

Someone asked me recently how to wring every single last little drop of revenue from their website. So it got me thinking about every page on a website and how it could be monetised.

There was the usual ‘hey, we could put an AdSense Ad on this page’ and an ‘Affiliate link on that page’ etc... But then it came to the final conversion page...

What could we do with that?

As it turns out, monetising the conversion page has made him more money than any of the other page on his website he’s tried to monetize.

Let me explain.

The final confirmation page shown to someone after they’ve purchased from you is actually quite a powerful sales page.

The reason being is that someone who has just completed a purchase from you is definitely in a buying mindset; they’re probably on a bit of a high and ready to get their next consumer fix.

But conversion pages are typically a notorious let down! You get through an exciting shopping and checkout process only to be confronted with, “Thanks for ordering from us” at the very end.

What a disappointment!

They’re almost always ignored and very little, if any effort goes into them.

The only half decent one I’ve seen of late is Deals Direct’s which has quite a bit of guff about shipping, guarantees, live help etc. Its ok, but I reckon the only reason it’s so full of information is so that users stick around for a few seconds more than normal so all their A/B split testing script loads properly and they can get some reliable data on conversions! (Sorry, I’m a notorious ‘page source snoop’).

But back to the often ignored conversion page and sales...

We figured that once someone had bought something from the site, they probably weren’t going to come back for a while to buy from us again. In fact the chances are slimmer than Lindsay Lohan after rehab.

So we signed up to another ‘complimentary’ retailer affiliate program. Not a direct competitor, but an online retailer who sold goods that complemented the ones being sold on my friend’s site.

It worked!

Lots of clicks off to the other online retailer and quite a few sales generated as a result! (about $100 in a month worth of commissions and a conversion rate over 8%!).

OK so he’s not bringing forward his retirement plans, but an extra $1,200 per year gets him half a holiday.

You could try this if you sold laptop computers, but not laptop bags for example. Or if you sold flowers, but not chocolates; holidays, but not guidebooks; cameras, but not photography courses... the list is almost endless...

Update; 26 Feb 2010

My colleague Liesl Pfeffer came across another really cool checkout - marketing technique using a "tweet this" button (in the bottom right hand side of the page). Watch the sequence.

What factors can influence conversion tests?

Improving your website conversion rate is something everyone should be constantly working on.

Obviously, lots of people in our industry bang on about the importance of A/B split testing and Multivariate testing as tools to lift Conversion rates. And more often than not, these systems work.

But there are factors which can come into play which can knock your testing efforts for 6.

So here are 6 factors which might be affecting your conversion rates and your ability to test effectively.

1. traffic spikes from left field.

We recently had a situation where a client received a huge amount of junk traffic, right in the middle of a test. The traffic came from a new affiliate the merchant allowed to promote her site called AdFlasher – basically a system where people get paid to click on affiliate banners on the AdFlasher site.

The bounce rate from this traffic was sitting at 99% the reason being is that people actually get paid to click on as many banners as they can in a 10 second time frame. Can’t for the life of me work out how AdFlasher makes any money – none of the traffic was converting into sales - but traffic went through the roof regardless. By the time we’d figured out what was going on the test lay in tatters.

Always test the integrity of your traffic!

2. Conversion ‘pressure’

Conversion pressure is a phrase I’ve coined to describe a phenomenon which rolls around every year. Quite simply, it’s the pre-Christmas shopping rush.

Nearly every retail client we have sees a lift in their conversion rates in the weeks leading up to Christmas as online shoppers make last-minute, hasty and impulsive decisions to buy gifts. They’re under pressure and it shows! Conversion rates can double without lifting a finger, but don’t expect it to last post Christmas. It never does.

3. Seasonality (needs)

Like conversion pressure, seasonality also can affect your conversion rates -depending on your industry. Usually B2B sites experience abysmal conversion rates over summer. People are visiting their sites, but no one’s filling in enquiry forms! Folks in business are looking and planning, but they’re usually on leave.

If you’re selling sunglasses – then summer will probably be a good time for conversion rates. A bit like running a carwash after a dust storm. Same amount of cars driving past, but more people driving in to get their car washed.

This is why it’s always important to be testing all the time (Always Be Testing!).

4. Competitors

Competitors can definitely affect your conversion testing efforts. A big sale on a competitor site will probably sink your efforts. In many categories people do browse quite a bit online looking for the best deals, which means if a competitor does something dramatic in the middle of your test, the effort is wasted. Pull the test and compete!

5. Currency

If you sell internationally, then you might be seeing your conversion rates starting to drop, especially if you advertise your prices in international currencies, mainly because the Australian dollar is strengthening rapidly against most other currencies.

Same amount of traffic, but fewer sales due to the fact you’re less competitive on the pricing front. This also plays havoc with your testing. And don’t forget, it can work the other way too if the currency weakens.

6. Annualised tests

Don’t run a test just before Christmas this year with the intention of running it again the same time next year thinking you’re going to get conclusive results. You can’t, because so much can change which is outside your control. The variables are just too great, your data will be different, economic circumstances will have changed, and so on.

I mentioned earlier, Always Be Testing, so run your tests as quickly as possible which and check the integrity of your data during and after your test.

If it all sounds too hard, don’t forget to stick to the basic tests that always work; keep up your usability task-based testing (try http://www.usertesting.com) and run http://www.Crazyegg.com and http://www.Clicktale.com over the site once in a while!

Go Saints.

Ten steps to higher conversions using Google’s Website Optimiser

This year the really big trend by online businesses in Australia has been towards landing page optimisation and testing. As usual, we’re about 12 months behind the US and Europe, who have always been the ‘early adopters’.

For the big offshore sites, their SEO strategies have matured; they’re ranking highly for targeted traffic and their brand-related keywords. Naturally, the next step is to convert as much of that traffic into revenue as possible.

It’s also vital to optimize and test landing pages to improve conversions given the constant changes to Google’s AdWords system. It’s becoming more and more expensive to bid on keywords, so improving your chances of high ROI makes perfect sense.

Landing page testing tools have been quite expensive, putting off the smaller players. However, Google has created a free tool which can make testing your pages inexpensive; the only expense is your time.

If you’ve got a Google Analytics or Google AdWords account, you might have noticed a link to a free product Google’s offering called Google Website Optimiser.

In Google Analytics you’ll find it in the “visitors” section



In Google AdWords you’ll see it as a tab…



Note: You will need Google Analyics to use the Website Optimizer tool.

If you have basic HTML skills and have access to pages on your website it’s not too hard to set up, so let’s go through the steps together. This will really help your online sales!

Step 1. Click on the Website Optimizer link.

Glad we got that out of way.

Step 2. Create your experiment.

Select which one you’d like to try – A/B split testing is the easiest, but Mutivariate testing is probably the most powerful.

I’m going to focus on Multivariate testing today because many of you have product pages in shopping carts which will really benefit from this.


Google says you should have around 1,000 page views a week, but it will work on pages with much less than that, you just have to wait longer for the results to shine through. So treat this as a long term project.

Note: there are more expensive 3rd party tools on the market which can speed reporting up considerably which use the Taguchi method of statistical analysis.

Step 3. Choose the page you would like to test

You can test your homepage or a product detail page for example.

Step 4. Choose which elements of the page you’d like to test

Examples of potential test page sections could be headlines, images, testimonials, call to action buttons and promotional body copy.

Step 5. Identify your conversion page(s)

This is an existing page on your website which users reach after they've completed a successful conversion. For example, this might be the page displayed after a user completes a purchase, signs up for a newsletter, or fills out a contact form.

Step 6. Input the page addresses into the forms

A screen shot tells a thousand words. This is very easy!



Step 7. Input the test codes

Google supplies you with special code to place into the testing page. Again, at first glance this appears complicated, but it’s really just a common sense ‘cut and paste job’. You’ll also need to add code to the conversion page.


Step 8. Add the variations

This is easy too. Just enter variations for Headings, Images, Body Copy, Buttons etc. Try not to go too far and add scores of variations. It will make your experiments far too complex and you’ll wait even longer for meaningful results to be established. So add 2 – 3 variations for each element.

Step 9. Preview and Launch

Here you can double check everything you’ve done and then hit the 'launch' button.

Again, nothing scary here.

Step 10. Read your reports and take action!

This is the fun part where all your hard work ‘pays off’. If you’re running an ecommerce site for example, you may have found conversions increased by 5% using a certain combination of images and messages. I’ve heard of instances where conversions have jumped as high as 20%!


You can roll out the successful results from your test out the rest of your product pages, potentially creating a huge uplift in sales and revenue across the board.

As they say, times are tough, so give it a try – it might help you ride out the down-turn while your competitors whither under the weight slowing sales.