Want to improve the relevance of your landing page without changing so much as a period? Improve the quality of your paid search ad.
I’ll admit there doesn’t seem a direct causal connection between the perceived relevance of your landing page and the quality of your search ad. That’s why Microsoft hired Enquiro Research for a project earlier this year to better understand the impact of ad quality. Enquiro used eye tracking to directly measure user engagement. Turns out that the interaction between searcher, search ad and landing page is pretty complex.
In 8 seconds I’ll either be gone or give you
the opportunity to sell me on the next click.
Let’s paint the picture. You have 8 seconds or less to convince me that your landing page is relevant to my interests. In 8 seconds I’ll have made up my mind. In 8 seconds I’ll either be gone or give you the opportunity to sell me on the next click.
That’s not a lot of time to make a conscious decision. In fact, a lot of the decision I’m making isn’t conscious at all. It takes place under the hood. Sparks are igniting and pistons firing beyond my awareness or control despite the fact that I think I’m steering the car. Enquiro’s research suggests that one of the clues I’m using to make my decision about your landing page is the quality of your ad. The aftertaste of a bad ad is lower perceived relevance of your landing page.
Even a bad ad can be good enough to get a click sometimes. In fact, really bad ads can generate more traffic than really good ads, it’s just not the kind of traffic you want. There’s little return on your advertising budget. So what makes a bad ad? You’re likely to recognize one immediately - if it’s not your own. A bad ad isn’t specific. It doesn’t speak directly to the searcher’s intent. It’s vague and unconvincing. And often it’s the result of an advertiser trying to cover too many diverse keywords with the same generic copy.
A bad ad will make a good landing page worse
but a good ad has less ability to make a bad page better.
So what’s a good ad?
- Clear, specific, concise
- Relevant to the searcher’s goals
- Reflects the searcher’s keywords in sensible ad copy
- Communicates the landing page’s value proposition
- Incorporates a strong call to action
A bad ad will make a good landing page worse but a good ad has less ability to make a bad page better. It’s one of those life lessons we all learn. Bad news has longer legs and travels faster. That makes it even more important to marry a good ad with a good landing page. But what describes a good landing page?
- Your visitor arrives at the promised content or with one, but no more than two, clicks
- Your content is unique and informative
- Your content is directly relevant to the searcher’s keywords
- Your site has some reason for existence besides advertising
What can you do to identify a bad ad? There’s a simple, nearly foolproof metric: bounce rate. Well, it would be simple if every publisher of web analytics software measured bounce rate the same way.
Identify where your search advertising budget is hemorrhaging.
Bounce rate can be defined either as a) only one page viewed by a visitor in a single visit, no matter how long the visit, or b) only one page viewed by a visitor in a visit that lasts typically 10 seconds or less. It’s the second metric we want to track.
Segmenting your traffic by search engine campaign, ad group, and ad can help you identify where your search advertising budget is hemorrhaging. Each click is costing you money, each bounce is a lost opportunity. Once you identify your worst ads, fix them.
- Rewrite you ad copy
- Use negative keywords
- Reorganize ad groups with keywords having tightly related meanings
- Use geographic and demographic targeting, when appropriate
- Use dayparting (with the same caveat)
Monitor your bounce rate per ad. If the bounce rate declines, keep the change. If not, discard it. Develop a hypothesis to better understand both. Test your hypothesis against your bounce rate to see if you’re right.
You’re not only learning what works in search advertising, you’re practicing the scientific method!