There’s a lot of debate over how much titles and especially meta descriptions affect SEO. Here’s a definitive way of figuring out whether adjusting titles and meta descriptions is worth it.
This blog is provided for general information only. Always talk to a laywer or other appropriate professional before making decisions.
What do titles and meta descriptions do for your SEO?
To understand how to analyze titles and meta descriptions, it’s important to understand what they do. They have two jobs.
The first job is to tell search engines what a page is about. This job isn’t as important as it used to be. People figured out how to game the system, so search engines figure out how to read the entire page. Your title and meta description should still say what the page is about.
The second job of your title and meta description is to get people to click your page. Think of the search results page as a shopping center. Once people arrive, it’s the store signs and window ads that convince people to walk in.
Now, you might be saying, people only drove there because they’re looking for a specific store. In some cases, that’s not true. They might decide to go to a mall but not which store until they get there. In other cases, they might be coming to you no matter what.
How do search engines rank your pages?
Search engines use many factors to rank your pages. A lot of it is the content on your site and links to your site.
Part of it is how many people click on your website when they see it in search results. A click means people think it will probably be useful. More clicks might equal increased search rankings.
But you don’t want to trick people into clicking on your website. Another important factor is the bounce rate. When people quickly leave your website and go somewhere else, it tells search engines your site probably wasn’t useful. That makes your search rankings go down.
How do you know what ranking factors to focus on?
So how do you know what your website needs to rank higher? Is it more content, more links, or better titles and descriptions so more people click?
To decide whether to focus on titles and meta descriptions, look at click-through-ratio. Click-through ratio or CTR is what percent of people that make a search click your site. For example, if your site appears in 100 searches and people click it 10 times, your CTR is 10%.
How do you know if you have a good click-through-ratio?
There are two ways to find out if you have a good CTR.
The first is by looking at averages. You’ll want to find the average CTR by search position. The #1 search result usually gets clicked 34% of the time. The #2 search result gets clicked half as much at about 17%. By the #10 result, it’s down to about 3%.
These averages will vary by industry and even search terms within an industry. In some searches, people are more likely to click the first result. In others, people are more likely to read through the results or click multiple pages. It’s much harder to find this data.
Once you have the averages, they’re pretty easy to use. If your CTR is above average, your titles and meta descriptions are probably working well. If your CTR is below average, you might want to make some changes. For example, if your Vacuum Cleaner Sales page ranks #10 and has a 1% CTR (below average), you might want to make improvements.
The second way to determine whether you have a good CTR is to look at your own stats. For example, say you have five pages that are ranked #5 in search results. Their CTRs are 10%, 7%, 5%, 3%, and 2%. You’ll probably want to adjust the 2% and 3% CTR pages while leaving the higher ones alone.
How do you find your click through ratios?
You can see your CTRs in your performance results. Some people will eyeball those reports to make changes. Other people like to export to a spreadsheet.
If you’re a spreadsheet wizard, you can use the linear forecast function to show your predicted CTR for a given search position. (Hint X=CTR, Y=Position). Since search results are logarithmic, you might want to adjust that function to use a logarithmic curve instead of a straight slope.
How do you know if you should adjust your titles or meta descriptions?
It’s harder to know if you should adjust your titles or meta descriptions since both affect your CTR. There aren’t really any stats to compare them directly.
If you don’t have good meta descriptions, start there. If you have meta descriptions but think your titles are weak, change the titles.
If you don’t have any clear answers, it’s time to do A/B testing. Create a few different titles, use each one for a few weeks, move on to the next one, then pick the one that got the best CTR. Do the same thing with meta descriptions when you’re done.
Don’t change both titles and descriptions at the same time. It’s like a science experiment. If you change two things at the same time, you won’t know which thing got the results. For example, you might change both your title and meta description and get a higher CTR. But, it might be true that you would have gotten even higher results with the new title and old meta description.
What adjustments should you make?
It depends on what you’re trying to get results for. Clickbait titles do well in some searches but poorly in others. Sometimes you want to phrase things as questions, other times, you want to present direct answers. A call to action sometimes works in meta descriptions, while other times, you want to provide information.
Since you’re probably targeting a variety of searches, you’ll want to experiment on individual pages. What works on one might predict what will work on a similar page, but there are no guarantees. It’s a continuing process of trial and error.
How long should you wait between making adjustments?
When you make adjustments, you need to get enough data. Think back to statistical significance from your high school or college math.
You’ll probably want to run each change for at least 1,000 search impressions and maybe 10,000 or more to get good data. How long that takes depends on the search volume for the terms you rank for.
You’ll also want to make sure you’re accounting for seasonal variations. If your industry has busy or slow times, don’t test option A in a busy time and option B in a slow time. B’s stats might be worse because of the slow season, but B could actually be better than A.
Are there any SEO title and meta description A/B testing tools?
Unfortunately, there aren’t many tools that can help automate this. Depending on the size of your business, this is often something you hire an SEO agency for. If you’re smaller, it will be a manual process. However, since these small changes can potentially double how many people click on your website, it’s a process that’s often worth doing.