What is Click-Through Rate (CTR)? CTR Optimization Tips

Actionable tips on title tags, meta descriptions, and rich snippets to boost search click-through rates.

Speakers: Ezgi Gülsen Yaylı

01 Jun 2017

Video Transcript

⏱ 8 min read• 1,381 words

Read the full speech or switch languages

Hey everyone. In this video I'll talk about CTR, click-through rate. What it means, why it matters, and how to optimize it. Let's start with the definition. CTR is clicks divided by impressions. So what does that mean? Your page gets impressions on a query. How many of those turn into clicks? That ratio is your click-through rate. Why does it matter? When your CTR goes up, your traffic goes up too. And when those visitors behave well on your site, that reads as a signal. That signal can push your rankings up.

There's one thing about CTR math that can get confusing lately. So I want to flag it here. Say you rank number one. There are still features like the Knowledge Graph and answer boxes. Those Search results page features sit above you. They often already answer the user's query. That naturally drags your CTR down. Why? Even at number one, something can sit above me. Call it position zero. That page grabs the clicks instead. So my CTR looks like a number two or three ranking.

So where do I get this CTR data? The main spot is Search Analytics in Search Console. Almost everyone uses it. Under Search Analytics you get queries and pages. You also get device, Search type, Search appearance, and date. I can filter the data by these. When we calculate click-through rate, we mostly use a few metrics. Query, page, device, and date. For date, the furthest back we can go is 90 days. On device, mobile and desktop users have different expectations. Their behavior can differ too. Also, the values that drive our click-through rate get different pixel limits. Things like Title and Meta Description. So splitting by device is pretty useful.

Pages and queries are what we really care about. After these filters, the page shows a few numbers. Total clicks, total impressions, average click-through rate, and average position. Right below that there's a chart. It plots the change over the date range we picked. After that, I wanted to show the view with pages selected. I click Pages, and my pages are listed here. You can see clicks, impressions, click-through rates, and average positions. If we export data here, it gives us 1,000 rows. But you can get more with other tools. Like I said, what actually drives click-through rate? What makes a user decide to click or not? It's the visible parts of our page on the Search results pages.

So it's the Title. The URL. The Meta Description. And if I've done structured data, the parts that show up. Title and Meta Description are where I describe the page content. That's where I tell users what they'll find inside. They read it right on the Search results page. The URL is just that page's URL. Alongside it, you get the image I'm showing here. You might also see user ratings or reviews. For recipes, you might see cook time in minutes. Not every markup tag we use shows up here. But we run into the main visible ones all the time.

I've covered why this matters. I've covered which metrics I'll optimize. Now let's get to the optimization part. First, I need to collect data. Like I mentioned, I can already pull this data from Search Analytics. Back here. When we look at click-through rate, what do I need? Really, what do I need to collect to raise it? Which queries land on my page? Here's the Search Analytics feature we'll use most in this work. I pick a single page right here. Say /click-through-rate. I click that page. Then I select queries on the page that loads. Now I only see queries where this one page performs in Search results. Just the ones it gets impressions or clicks from. So what else can I do after that?

I can test those titles on AdWords or Facebook. Then I look at the click-through rate there. So I run an AdWords or Facebook test first. That comes before testing organically on the Search results page. After that, I use the data I got. That's where those 4 metrics I mentioned get optimized. First, I gave examples for the title. What should a title look like? Quick list: descriptive, striking, with numbers, with adjectives. Here's a descriptive one. "What is click-through rate? What can you do to improve it?" Someone searching reads that title. They instantly know what's on the page. They know what the content holds.

Or striking. "34% organic traffic lift in 3 weeks: click-through rate optimization." That one really lands. Say "34% organic traffic lift" and few people skip the click. If the content fits, work a number in. Like "14 click-through rate optimization tips." When I hit that page, I know what I'm getting. 14 different tips. Or use an adjective. "The best click-through rate optimization tips." "The easiest cake recipes." Next up is our URL. The URL isn't something people test much. Say you use a URL like domain.com/346.html. Someone looking at just that URL has no clue. They can't tell what's on the page. But make it descriptive instead. Something like domain.com/category-name/product-name. Now they know they'll land on the product they wanted. And it sits under the right category.

Then third, you've got structured data markup. I mentioned some metrics earlier. You see these labels here. What could they be? Say I'm shopping for a product. If it says 34 reviews, I want to click in. I want to see what actual users think about it. That pushes me to click. Or the user rating is 4 out of 5. I think, okay, this product is actually good. Again, that pushes me to click. Same with images. I can see the product I want is right on that page. Yes, that's exactly what I was looking for.

Take the example I gave earlier. Someone searches for the easiest cake recipe. They search for easy cake recipes. Then they see 20 minutes here. They think, okay, a recipe I can make in 20 minutes. That really is easy. Then fourth, you've got the Meta Description. With the Meta Description, I explain what the page is about. Why? I get a bit more room here. Google gives me more pixel space than the title. The Meta Description used to count as a ranking factor. If I remember right, that was back around 2013.

It's not a ranking factor anymore. But it still lifts your CTR. Then I get to the third stage of optimization. That's implementation. Now I'm setting different values. I took those 4 metrics I mentioned. I gathered my data. I looked at my queries. What did I say? I looked at the queries getting clicks or impressions. Then I spotted keywords with a low click-through rate. My page got plenty of impressions. But the click-through rate is low. I might want to work those into the title or meta description.

But I really need to set different values. Don't just reshuffle the existing title and put it back. You need to test genuinely different words. Base them on those three metrics I mentioned. And while testing, don't change things too often. So not: I put one in, then tried the second a week later. You have to let that title gather some data. Why? If I test three titles a week apart over three weeks. Positions may shift in the meantime. That position shift could push the click-through rate up or down. And I can't credit that to the title I tested. So I shouldn't change it too often. I need clean data. How often, then? That depends a bit on the page's target keyword. Its Search volume or its position.

Once the data comes in and adds up, we can tell. Whether this title helped me or hurt me. After that, I need to start measuring. Neatly: when did I put this title up? When did I start the test? I need solid timestamps. And then I change the next metrics. After that, I run tests on those 4 metrics I mentioned. Then I take the best performer. And I stick with it forever. With a high click-through rate. That's all I had for today. See you in another video.

Take care.

This transcript was generated using AI speech recognition and verified against subtitle style rules.

Ezgi Gülsen Yaylı

Content author

Ezgi Gülsen Yaylı

both

Contact us

How we handle your data

Open as a full page