Let me tell you a story about how I used two easy landing page optimization tips to reduce my cost per conversion by $10.67 and improve my quality score at the same time.
Shorter Forms Reduced Cost Per Conversion by $10.67
As you know, Marketo provides on-demand marketing software including lead management and landing page optimization. One of my jobs as VP Marketing is to generate qualified leads for the sales team by finding prospects using Google AdWords (as well as other channels) and nurturing them until they are qualified leads. In this sense, my job is similar to that of any other business-to-business marketer.
I have a variety of ad groups, but let’s focus on the one that talks about Marketo’s ability to create landing pages with no code or IT involvement (a feature of our lead management solution). What’s special about this ad group is that I’m using Marketo’s landing page testing capabilities to send the clicks to one of three different landing pages. Here’s a screenshot of one of the three:
The only difference between the three pages is the form. As we’ve all heard before, simple forms convert better, but the question is, how much better? Is the extra information worth the reduced conversion rate? Are certain types of information more costly in terms of conversion rate? I decided to use landing page testing to find out.
First, I had to determine how many test versions I could have. As I wrote in Landing Page Testing: How Much Is Too Much?, it’s easy to create tests that are too complicated to achieve statistical significance in a timely fashion. So, I used Marketo’s Landing Page Test Calculator to calculate that this ad group could support three test versions in the time I wanted to run the test.
Next, I used Marketo to create three different forms, appropriately named “short form” with five fields, “medium form” with seven fields, and “long form” with nine fields, and used these on three different versions of my landing page.
Finally, I let the system automatically rotate between the three versions and report the results in terms of conversions. Here’s what I found:
- Short Form: Conversion rate 13.4%, cost per conversion $31.24.
- Medium Form: Conversion rate 12.0%, cost per conversion $34.94.
- Long Form: Conversion rate 10.0%, cost per conversion $41.90.
The difference in the cost per conversion between the short and medium forms is $3.70, between the medium and long forms is $6.96, and between the short and long forms is $10.67. This means that each additional piece of information costs $1.85, and that asking for a phone number (the most invasive of all the fields) costs more than $5.00. At these prices, the conclusion was obvious: keep only the short form and turn off the other versions.
As much as my sales team and I would like that information, it is not worth paying for these extra fields. Fortunately, this type of information is available from a variety of data cleansing and augmentation vendors, usually for as little as $1.00 for all the information you need.
Landing Page Metadata Improved Quality Score
Another question I wanted to test was “What impact does the metadata of the landing page have on determining quality score”? To figure this out, I used my “lead management” ad group. To begin, I created a targeted ad and landing page that focused on the concept of lead management and offered a free lead management eBook. A sample text ad was:
Free Lead Management Tips
10 Tips to Optimize Sales Leads
Download Marketo’s free eBook now.
Here’s what the landing page looked like:
Clearly, this page is all about lead management, but I left the page metadata (e.g. title, keywords, and description tags) generic. When this ad group was brand new, Google assigned the following quality scores:
Next, I updated the page metadata (e.g. set the title tag to “Lead Management Best Practices”) but did not change any other element of the ad group or landing page. And you know what? It turns out just this small change had a real impact on the quality scores:
For most words, the minimum bids went from $0.15 or $0.20 down to $0.10. Interesting, the only word that didn’t get a quality score improvement was “prospect management”, which makes sense since I didn’t include that term in the title or description tags for the page. Perhaps that term would work better in its own ad group.
The implication of this is that each ad group should have its own landing page, specifically targeted to the keywords in that ad group. Even if the content on the page is the same or similar, just tweaking the metadata can have a positive impact on quality score.
Note: You can read a longer version of this in my recent article at Search Engine Land.