How to Improve Title Tag CTR by 20%+

By | Last Updated: | SEO

Title Tags for SEO

One of my favorite new actions for every new client we take on at Siege is analyzing title tags. There is still a glut of websites who are using 2005 title tag best practices, costing them significant profit in the process.

Given what we know about over-optimization, Panda and the added potential that lower CTR could also potentially cause a reranking – and better CTR a reranking for the positive – it makes even more sense to think extremely deliberately about our title tags and modify them away from our old-school titles towards a better set that can immediately pay dividends.

In doing these tests for our clients, we’ve often experienced 20% or better CTR improvement for titles that have that “2005” feel, across a large sample size of markets and pages. Some are of course better than others, but when you’re starting from square one, there is significant room for improvement that may be available for you as well.

Start From Zero: What Are Title Tags?

Before we go too far, let’s establish the basics in case you’re new to the SEO game. Title tags, technically known as title elements, create the title text you see in every tab of your browser, and importantly, in the search results you see every day.

These titles, created with HTML code in your header, are meant to summarize your page. In historic times, it was thought that the closer to the front your keyword was in these titles, the better you would rank. This is still true to a degree, but it’s increasingly the case that Google knows you are doing this for SEO, and hurt you because of it.

We’ll suggest ways to get around that in this post.

Title tags look like this in the header of your source code:

  • <title>Title Tags: How to Improve CTR by 20%+ – Siege Media</title>

And like this in the search results:

Screen Shot 2016-03-10 at 9.09.45 PM

In essence, they should be tailored around the keyword focus of your page, aim to be shorter than 55 characters, and also incite users to click — which we can do through the tests we will discuss in the paragraphs to come.

Best Practices Behind CTR Tests

The general philosophy behind these tests is that there is no longer need to shove 90 keywords in your titles, and in fact having those keywords may actually prevent you from ranking more holistically.

Therefore, by removing long tail iterations we can have confidence Google should be able to figure out what you’re relevant for, and instead adding irrelevant ad copy to improve CTR, we can create a net overall lift in traffic to the site.

Once we understand that, we can then immediately gleam CTR best practices from competitors (or ourselves) bidding on keywords on the search results. If we’re in any kind of competitive market, these title tags have been labored over and A/B tested into infinity, which makes stealing their information for our uses a no-brainer.

If we compare these tests over our old school SEO title tags, there is massive room for improvement. Many companies are still stuck in the past—writing title tags that are getting rewritten or don’t use any ad copy fundamentals—and therefore are likely leaving 20% or more improvement in their website traffic on the table.

Creating a Relevant Test

While often times you can just swap out title tags using this philosophy and see a big CTR change, it’s still worth setting up a testing environment to establish confidence the best you can.

Because true A/B testing with organic results is impossible, we can’t do a truly accurate test, but we can set ourselves up to gain confidence from the changes we make by evaluating the right factors and measuring our changes against them.

In Webmaster Tools, we can track average position and also CTR for our pages for a given date range in the “Search Queries” section in the “Top pages” tab. We use pages versus queries because we need to look at the impacts on all keywords on the page, not just the head term.

Screen Shot 2015-01-06 at 11.07.43 AM

When doing a test, we should aim to look at a window of time (generally two weeks or more) that gets us 2000 or more impressions on the search results. The higher the number, the better.

Every gust of the wind can change our confidence in the test—from a holiday weekend to seasonality to ranking changes—so it’s important to be wary of variables like that when setting up tests.

When we change the title tags, we don’t need to note the CTR and average position at the start of the test, because Google Webmaster Tools tracks that historically, and we can find the data after the fact.

Depending on the scale of your tests and/or the importance of the pages, you may want to track in a few different ways. When we’ve wanted to test an important page with high traffic, it’s best to use Webmaster Tools.

Here, you want to take note of four variables:

  • Clicks
  • Average Position
  • Clickthrough Rate (CTR)
  • Impressions

With this, you’ll want to also match up similar date ranges. Not Monday to Wednesday and Thursday to Sunday—more likely, two week ranges if your page has decent traffic. Once you push the title tag live, you may want to take note of when you first see it in the SERPs as the “starting date” for the test.

Once you see it live, it’s time to wait. When two or more weeks have passed, go back and record the data for your test by customizing the date ranges in Google Webmaster Tools in the same area mentioned previously. Compare it against the previous date range in a spreadsheet to determine how each field changed.

Screen Shot 2015-01-08 at 3.14.09 PM

When you get your numbers back, it’s time to consider how confident you are that the changes benefitted you. Did your rankings increase? This is possible, hypothetically, if CTR impacts ranking, and/or your more natural title tag appeared more relevant to Google.

Secondarily, and more obvious, did your CTR and therefore traffic go up? If CTR is up considerably and ranking dropped slightly, it doesn’t matter. Relevant traffic is what matters, not an arbitrary ranking position. 

If your CTR jumped considerably and average position stayed relevantly flat, congrats, you’ve almost certainly made a successful change. If your average position jumped solidly and so did CTR, you might not be able to confidently determine your test was successful. That’s a first world problem, though—you have increased traffic regardless, so it’s not worth reverting the change.

One very important thing to do is have a “control” relative to your test page. We can’t have a true control, but we can compare how position and CTR changed for the site in Webmaster Tools, or other pages similar to the one we’re testing. This can help increase confidence in our changes when we see our test page outperforms the control.

Running Long-Tail Tests on Page Types

For pages without enough traffic to reasonably test, that still have a decent amount of traffic and a similar theme—such as profile pages on Moz—you can group them together to get more confidence in your changes and then test the impacts on traffic in Analytics. You can do this by creating an advanced search in GA for only those pages that will identify the traffic impact from organic across your date range.

You can create an advanced search by using the | symbol between URLs used under the Site Content -> All Pages section in Google Analytics. An example using two URLs is shown below.

Screen Shot 2015-01-05 at 1.56.41 PM

Once you’ve setup the segment for just those pages, you can compare date ranges from before and after you implemented the title tag change. If the improvement is big (it should be, if you do this properly), you can have confidence in your change. If it’s down, revert. If it’s flat, probably leave as is.

Screen Shot 2015-01-05 at 1.49.53 PM

If you do a good test/are working against old school SEO tactics, there’s a good shot you can see 20% or better CTR increases with a good title tag improvement. And that’s passive—I’ve seen pages experience 50%+ jumps depending on the SERP and the industry.

This is the perfect kind of test for making big improvements. If you’re trying to make incremental .5% increases with organic title tags, you’re in the wrong line of business—there are too many moving parts to really have confidence in your changes.

That doesn’t mean you shouldn’t ever make changes with CTR in mind, but it’s something to consider for future campaigns. You’re going to see the biggest gains with the first well-ran set of tests, especially if you haven’t done this before.

How to Generate Improvements in CTR

Having done enough of these tests to date, there have been a few patterns we’ve noticed about what works. The title tags contain one of five elements that drive clicks, and sometimes two to three of them combined.

title-tag-ctr-diagram

The Low Price Reference is as simple as it sounds. If we’re a low price leader in the market—or even competing for the position, we should be mentioning this price in the title tag.

This often comes in the form of noting the lowest price you have available across your product/service range. “Low price” can be a reference to things that aren’t directly cost related, too, such as with home mortgage rates.

Screen Shot 2014-11-12 at 8.35.20 PM

The Freshness Reference refers to showing how updated you are as a reason to drive clicks. It’s been mentioned this year how much Google favors freshness—it seems to be dialed down a bit recently but it only became a thing because users prefer it—which is obvious to me based on how often showing freshness in the form of dates (such as 2014) in the title tag can actually drive clicks.

The Volume Reference shows how comprehensive and impressive your content is by showing the sheer volume available if users click through. For example, if people were looking for things to do in Miami, it’s possible saying you had 122 attractions listed might attract a click.

 Screen Shot 2014-11-11 at 9.22.42 PM

The Speed Reference tells people interested in getting things done quickly that it will get done quickly based on your title tag. This could be a reference to shipping, how quickly you might get a degree, or in a space like PayDay Loans, how fast you can get that loan in your pocket.

The Brand Reference is the most powerful one of all, and the one that drives the most clicks in the least amount of characters. If you’re Geico, just those five letters will bring some extreme weight on a “car insurance” SERP that no combination of other keywords could.

When combined with great title ad copy, though, even Geico has the potential to drive more business to their website using one of the four other references.

Examples in Every Competitive Market

To give you a feel for what a recommended title tag change would be, showing examples of sites that could improve in every competitive SEO market is a good start. In our recent post listing 750 popular Subreddits, we referenced the ten industries that spent the most on PPC. They will be the ones we again utilize in this article.

In each industry, I will reference one site I think can improve for an extremely competitive keyword. This is not a guarantee that my recommendations will increase CTR. It is only a recommendation based on a strong confidence interval from having tested 100s of pages in the past two years.

KW #1: “Car Insurance” – Finance & Insurance 

Screen Shot 2014-11-11 at 9.49.48 PM

Progressive’s title tag is currently getting rewritten. They have a great brand, but according to Google, if you search for “car insurance”, all you get is “Auto Insurance – Progressive”. Extremely enticing.

Screen Shot 2014-11-12 at 10.30.15 PM

As you can see from Progressive’s site, it’s clear they have a similar price savings offering as GEICO—they can save people $500 on average. The ad team for them, at the time of this writing, shows that highlighting a low monthly price also helps CTR. By using this low price reference in organic, they can likely increase CTR on an extremely valuable #1 ranking.

GEICO is doing a great job of this, and is the only person in the market using the low price reference in organic to drive clicks on the first page. However, they aren’t using that monthly reference, which is clearly important to CTR given many people in the PPC market are using it.

Recommendation for Progressive:

  • Old Title Tag: Auto Insurance: Get an Online Car Insurance Rate | Progressive
  • New Title Tag: Auto Insurance: Car Insurance from $19/mo | Progressive
  • See the SERP: Car Insurance on Google

You’ll notice that I recommended dropping online and rate, both long tail keywords for them. It doesn’t matter. Almost certainly, adding the ad copy that drives clicks will more than make up for that drop—especially because Google cares less and less about keywords these days and more about the user—which will be rewarded even more with a great title tag.

KW #2: “Refrigerators” – Retailers & General Merchandise

In this industry, there isn’t a lot of differentiation, either in PPC or in the SERPs. Lowes includes an ad about 10-30% off. If we look at the shopping ads above the fold, we see that the starting price is $999, which means that if we have a decent number of products below that number, we have the opportunity to stand out.

Specifically, Lowes says they offer fridges from $399+ in their ad, which sounds like a nice price point differentiator which they could use on their on SERPs—a use of the low price reference.

Screen Shot 2014-12-24 at 8.58.29 AM

Lowes currently has a pretty basic title tag without much differentiation. The thing they’re using well, though, is their brand—they know the significance and that’s probably why they use it so early in the title tag. That doesn’t mean they can’t improve, though.

Recommendation for Lowes:

  • Old Title Tag: Refrigerators at Lowe’s: Counter Depth Refrigerators, French Door Refrigerator
  • New Title Tag: Refrigerators at Lowe’s: Shop Fridges From $399+
  • See the SERP: Refrigerators on Google

One counterpoint you might be thinking is “What about the targeting of Counter Depth and French Door?” That’s a good point, but that’s a bigger sign of optimization issues with the rest of Lowes’ site, not this page.

They should have a better landing page experience for Counter Depth and French Door than a general refrigerators page. They should solve for that and switch to the title tag optimized for clicks.

KW #3: “Things to Do in Miami” – Travel & Tourism

Looking at the SERPs for this keyword—and also thinking about the search intent for the term—the proper go-to is the volume reference. You don’t want to click a result and get five things to do, what you want is a massive list that allows you to filter by interest to allow you to find the perfect activity for you.

Screen Shot 2014-11-19 at 8.19.50 PM

TripAdvisor does a pretty solid job of this and are the only current ones using a volume reference, but their problem is length at a whopping 68 characters.

According to Moz, the ideal title tag length is somewhere around 55 characters—you don’t always need to go that short, but somewhere close makes it more likely you won’t get rewritten.

With volume references, a best practice is to ping dev to auto-update your title tag with the newest number of said references. If you truly have a high volume, most of the time the number will be constantly evolving, so it makes sense to have the updates be an automated process.

Recommendation for TripAdvisor:

  • Old Title Tag: Things to do in Miami: Check out 122 Miami Attractions – TripAdvisor (68 Characters)
  • New Title Tag: Things to do in Miami: 122 Miami Attractions – TripAdvisor (58 Characters)
  • See the SERP: Things to do in Miami on Google

KW #4: “How to Become an RN” – Jobs & Education

The main differentiator for this SERP, a unique one, is time—a perfect application of the speed reference. Students who want to become a nurse—or get an education period—often want to do so in the fastest time possible. This can be identified in the PPC listings, where some ads reference the time it takes to get one and even offer accelerated programs.

Screen Shot 2014-12-24 at 9.13.52 AM

The interesting part about this search result is that it has lots of very straightforward “How to” articles from schools. You’re not going to be able to have an “RN Degree” page that will convert people direct, you’ll have to soft sell the offering, which several schools do.

There are offerings like the above, which might drive CTR, but may not be realistic for schools to promote if that’s not a realistic time for their degree program. Just because one ad has something shorter than your offering doesn’t mean you can’t see a CTR improvement by including something longer in an organic title tag—we know that generally many people do ignore ads, much to Google’s chagrin.

Screen Shot 2014-12-24 at 9.35.27 AM

The first result, Jacksonville University, is a good example of this. They have a very basic title tag which has no differentiation among other results—besides clearly being from a University specializing in nursing degrees.

Recommendation for Jacksonville University:

  • Old Title Tag: How to Become a Registered Nurse – Steps to Become a Nurse
  • New Title Tag: How to Become a Registered Nurse in 24 Months – JacksonvilleU.com
  • See the SERP: How to Become an RN on Google

Jacksonville may have to adjust their landing page experience slightly to account for a change that emphasizes time, but otherwise I don’t see it causing them issues. On a SERP where each of the top four results all start with “How to Become a Registered Nurse” with no other differentiation, it could help them stand out and possibly even convert more potential nurses.

KW #5: “Contractors in San Diego” – Home & Garden

On SERPs like this, generally the concern most searchers have is whether or not the contractor is the best option for them out there. By referencing your comparison data using the volume reference, national companies can stand out in these search results and make it obvious that they can help consumers make an informed decision on their website.

Screen Shot 2014-12-24 at 9.52.48 AM

For example on Houzz, they highlight the ability to browse through 2,976 local San Diego contractors once you land on the page. You can also sort by review volume and best match, as well as subsegment by contractor type. Their title tag, however, is not as compelling, simply using the main keyword. Why stop there?

Recommendation for Houzz:

Because of limitations, there are some issues with this title tag as ideally we’d like to get “contractors” at the end. Here, though, I think the meaning still transfers and Houzz does have a nice brand in the space, so it’s worth highlighting their significance in the industry.

The main needs here are to get the volume reference, the brand and the comparison capabilities all in the title tag.

KW #6:  “Laptop Reviews” – Consumers & Computer Electronics

With laptops—and technology—one of the prominent value-adds is in freshness. You don’t care about 2013 laptop reviews—you might not even care about six months ago reviews—and for that reason, the freshness reference is a good thing to utilize as a method of showing you’re worth being clicked on.

PCMag, ironically, does nothing of the sort. They are quite outdated in how they show their reviews are relevant. For that reason, it’s likely they’ll lose search share to LaptopMag, which utilizes freshness in the form of a year to their advantage.

Screen Shot 2015-01-06 at 10.34.26 AM

It’s possible to even do better than a year. For example, in credit card reviews, it’s extremely common to see the season or even month referenced as a reason why their content is the most up-to-date and worth clicking.

Recommendations for PCMag.com:

  • Old Title Tag: Laptop Computers & Notebook Reviews | Laptops & Notebooks Review | PCMag.com (76 Characters)
  • New Title Tag: Laptop Reviews – Spring 2015’s Best Notebooks | PCMag.com (57 Characters)
  • See the SERP: Laptop Reviews on Google

It’s important to note that this shouldn’t be a false prophecy—you need to have Spring 2015 reviews, or you’ll likely experience a poor landing page experience which quite possibly can cause Panda issues when people bounce back to the search results when they find themselves unsatisfied with your results. Thankfully, PCMag actually does have up to date reviews, so this title tag will result in a solid landing page experience.

Screen Shot 2015-01-06 at 10.38.30 AM

KW #7: “Car Loan Rates” – Vehicles

For loans, the call to action is APR—what’s the lowest APR you can get in order to save the most money on your loan? So while not a traditional low price reference, the lower the percentage, the more likely you are to be interested.

However, there’s a balance—brands and big banks likely can’t offer an APR they can’t match due to legal reasons. So, their main CTA is going to have to be their brand power. But for leadgen players who offer comparison products, it’s very much a possibility. In this example, Bankrate. On their car loan page, they don’t immediately show the lowest rate. However, they do show averages.

Screen Shot 2014-12-31 at 3.38.34 PM

Bankrate could pull from their database to retrieve the current lowest rate, or if not comfortable, they could show the average. Given nobody is using the rate in organic results, this could increase CTR. However, players in PPC are leveraging the lowest possible rate (under 2%), so to fight fire with fire, they’d want to use a lower number, if possible.

Screen Shot 2014-12-31 at 3.41.29 PM

This is one of those instances where it sometimes can be difficult to beat the PPC players because of the legal implications and also the reality of dynamic ad targeting plus needing to meet SEO best practices. But for the same reasons, because some traditional banks can’t use CTR improvements like including APR, it gives websites like Bankrate an advantage they can leverage.

Recommendations for Bankrate.com:

  • Old Title Tag: Auto Financing: Compare Car Loan Rates & Get Auto Loan Advice
  • New Title Tag: Auto Financing: Compare Car Loan Rates From 2% – Bankrate
  • See the SERP: Car Loan Rates on Google

KW #8: “Web Hosting” – Internet & Telecom

Web hosting pricing is a race to the bottom and many of us, those frequent browsers of the web, are worse off for it. Despite that, if you work in hosting, you’d be remiss to ignore said race. You need to utilize the low price reference in order to drive more clicks to your website.

As we can determine from the PPC listings, low price is the call to action of the day—of the monthly variety. In order to determine the right price reference to use—monthly, daily, yearly, etc.—it’s best to consider how your industry prices. In hosting, people compare based on the monthly cost. If in your industry (such as mortgage) the point of reference is the 30-year interest rate then that’s the differentiator you should use as your low price reference.

Screen Shot 2015-01-06 at 10.51.57 AM

As mentioned, it’s the monthly reference here that makes the most sense. GoDaddy purports a $1/month hosting package in their sidebar, which means they should be able to leverage similar in the organic listings. Combined with their brand strength, there’s no reason they shouldn’t be ranking #1 over SiteGround.

Recommendations for GoDaddy:

  • Old Title Tag: Web Hosting | Lightning Fast Hosting & One Click Setup – GoDaddy
  • New Title Tag: Web Hosting from $1/mo & One Click Setup – GoDaddy
  • See the SERP: Web Hosting on Google

KW #9: “Business Cards” – Business & Industrial

In the business cards space, there’s a combination of the volume reference and the low price reference that’s needed. People want a good number of business cards, and they also want them for cheap. You can see this usage heavily in ad copy being used in prominent positions.

Screen Shot 2015-01-02 at 9.15.21 AM

And not surprisingly, this same smart ad copy is not being used by the players in top organic positions, likely costing them a good amount of traffic. Vistaprint themselves is the perfect example—ranking #1 but with no reference to the variables that are clearly working for them in ads.

One thing to be aware of in situations like this is sometimes advertisers may be more aggressive with pricing in an ad versus on an SEO landing page, so Vistaprint might not be quite as comfortable pushing the super low price in organic. So, I suggest using the “from” price below as it doesn’t guarantee the price, although it’s clear you can get cards at that rate on the site.

Recommendations for Vistaprint:

  • Old Title Tag: Business Cards – Customized & Affordable | Vistaprint
  • New Title Tag: Business Cards – 500 Customized Cards From $9.99 | Vistaprint
  • See the SERP: Business Cards on Google

KW #10: “Wedding Favors” – Occasions & Gifts

For the uninformed, wedding favors are those gifts you give to all your guests as a thank you for coming to your wedding. By nature, for those with big weddings, getting these affordably wherever possible is preferred. So, the low price reference is a positive and can be used easily—even though it isn’t currently in any organic SERP. Missed opportunity.

Recommendations for David’s Bridal

  • Old Title Tag: Personalized Wedding and Reception Favors – Davids Bridal
  • New Title Tag: Personalized Wedding Favors from 35¢ Each – Davids Bridal
  • See the SERPWedding Favors on Google

This is also a unique situation in the case of David’s Bridal. You might look at their title tag and think that the lost KW optimization of reception favors might hurt them. Decent point, but if you research the search volume for reception favors, it’s around 500 a month in total volume.

Compare this to wedding favors, which is around 50,000. The benefit of getting even slightly 1% more traffic from the core set of keywords would make up for the the loss. Also, as mentioned above, it’s possible the improvement could be 20% or more—sometimes as high as a 100% increase.

A Note on Meta Descriptions

You might have noticed here that I didn’t include anything on meta descriptions. While I think including meta descriptions is a good best practice, in my experience, no meta description test has really moved the needle in a CTR test.

Implement them, but don’t worry too much about running expansive tests.

Why Isn’t This Being Done More?

In general, improving title tag CTR in organic SERPs isn’t done nearly as often as it should. Much of that comes from SEOs being stuck in older practices of keyword optimization and not moving fast enough into the new world where optimization matters less as Google gets better and better with understanding topics.

Similarly, many SEOs are better technicians than they are traditional marketers—understanding what drives more clicks is often a different game than understanding user psychology.

The good “new world” SEOs, though, will understand improving CTR in organic results is the way to go. It can make quick, massive impacts to the bottom line and it can help with other metrics that may actually improve ranking, such as improving potential brand signals for your business.

  • rjonesx

    Damnit Ross, stop writing amazing content. The rest of us don’t have time to keep up (or at least pretend not to).

    • rosshudgens

      Thanks Russ, I actually haven’t written a post in three months, so I sorta feel the same way..

    • http://officetricks.com/ kumarapush

      🙂 . Agree. Setting New standards every time. Hard for us to reach that level.

  • http://www.hoodwebmanagement.com/ Kane Jamison

    I thought you would bring up the | vs – from your tweet last week.

    As a personal addition, I love the use of ® on SERPs that don’t have many of them. For the State Farm example, it doesn’t add much because it’s hyper competitive with big companies already using trademark symbols. But for smaller niches where it’s uncommon, especially B2B or finance where trust is important, I think it adds a subtle but strong trust effect.

    • rosshudgens

      Kane I thought about that but after further deliberation decided not to. It feels somewhat coinflippy, someone gave me the counter argument that the – “connects” two sides of the title tag, which you can see explained here: http://searchengineland.com/9-seo-quirks-you-should-be-aware-of-146465

      Based on that I actually recommended to not switch for a few clients. Overall it’s a marginal thing so not something I think most should lose sleep over.

      I like your idea for the ® as something to test. I think I’ve actually noticed you did that yourself. 😉 Did you see a decent impact?

      • http://www.hoodwebmanagement.com/ Kane Jamison

        FWIW I hate “|” pipes in title tags, and I love hypens. Pipes always look like an SEO was involved and Google displays them in such an ugly way, and hyphens look way more pleasant to me. But, your tweet was a good point regarding pixel widths, so it’s now a tradeoff more than a style consideration.

        We primarily filed our trademark because it was the only way to get the @contentharmony Twitter handle. I added it in a few places (Twitter + our website) to support that filing process as well as for the SERP effect. AimClear does it on their site and I always liked how it looked. I have not been measuring its impact, so can’t give any data to support it, just my hunch that it is a trust builder.

        • http://www.tone.co.uk/ Anthony

          I’ve now implemented the ® for one of my title tag tests. Thanks for the heads up on that Kane!

  • theimaginaire

    Really enjoyable piece, Ross. Enhancing CTR with more effective title tags is one of the things that the majority of SEOs neglect that, as you rightly point out, can give a huge uplift in traffic. I’d be interested to hear how you’d approach title tags for premium products that don’t have an interest in being the cheapest?

    • rosshudgens

      Hey there, that’s really the hardest vertical/area to do much with. In those instances, it’s really a battle of brand. If you aggregate products you can sell a discount, which can still be a high #, like Bluefly does here: https://www.google.com/search?pws=0&q=luxury+handbags&gws_rd=ssl

      • theimaginaire

        Yeah, always interesting to get another opinion. I’ve found getting the usp of the client or the value they add into the title to be pretty effective for luxury products and services – especially in the business to business sector.

  • http://91digital.net Conrad O’Connell

    I was a little doubtful opening this given all the tweets about it, but the praise is well earned. Great post.

    Also a little note on separators, I personally love the bullet:

    Keyword Rich Title Tag Here • Brand Name

    Because of the visual punch in brings and (IMO) makes the brand name really stand out. The bullet doesn’t look like a letter like the pipe and has a nice visual appeal to it. I first saw this on Yoast’s website and now it’s my default brand separator in title tags.

    • rosshudgens

      Nice Conrad, I really like that too, if only because it stylistically looks good. Not totally sure it’ll help CTR but if it’s breakeven at least my title tags will good. 🙂

    • Georgi Bogdanov

      Hi Conrad can you tell what kind of symbols (except the “•” ) you’ve tried to implement to a title tag, because recently I tried to put “✔” symbol, but it seems that Google is not visualizing it.

      • http://91digital.net Conrad O’Connell

        Hey Georgi,

        The options in Yoast have all worked for me — see here: http://d.pr/i/13Ysr

  • http://www.problogger.gr/ Pano Kondoyiannis

    In the same position in SERP is increased by 53% clicks, 42% CTR and 7% impressions. That is excellent work.

    • rosshudgens

      Pano it should be noted that was a hypothetical test. However, it is representative of other similar tests I have ran in the past.

  • http://elnostreraco.com/blog/ @jrosell

    When comparing percentatges and you only change one thing, you can use chi-squared test.Look evan miller tool. I use http://www.evanmiller.org/ab-testing/chi-squared.html

  • Jessic7

    This is a huge and Great post 🙂 But I think you need to have a lot of courage to perform this test on your website. While changing the title tags of your important pages, There is a huge risk involved you may get the benefit but you may also loose your existing traffic and ranking in google at the same time. But I will surely try this one on my website.

    • rosshudgens

      I think this is greatly overrated, if you do it right. If you don’t feel confident in implementing the right way, then yes, you should have some concerns, but that’s where hiring consultants comes in. 🙂

  • http://www.tone.co.uk/ Anthony

    Hey Ross. Another top notch piece of content, excellent work. This has given me the kick up the backside to audit my clients’ title tags. I’ve already found numerous title tags that could and should be improved upon using your approach.

    One thing to note is that when you change the title tag, keep in mind that Googlebot has to crawl and update it’s index with the new title tag so the change in the SERPS isn’t instantaneous. Be careful what date you put down as the test start date. Ross, is it a case of just checking a couple of times a dayto see if the page has been updated in the SERPS or is there a quicker solution?

    • rosshudgens

      Hey Anthony, that’s pretty much it – another unscientific part of this. There are multiple data centers too, so it’s likely not everyone will get your TT at once. You’ll have to make an educated guess and just keep manually checking the SERPs in order to know when it’s live, and then use that as the “start” point of your test.

      As mentioned, you’re aiming for huge improvements here, not small ones, so hopefully even if we’re slightly off scientifically with a start date, we can make up for it with confidence.

      • http://www.tone.co.uk/ Anthony

        Thanks Ross. I’m confident I will see big improvements after following the advice in your post.

  • Chris

    Good post, I got some valuable information out of it that I will turn around for a client or two and my own site immediately. Thanks!

  • Benjamin Beck

    Hi Ross, Great post. Do you think that similar CTR increases could of been done with just improving the meta descriptions?

    • rosshudgens

      Great question Benjamin. Actually, when I started doing these tests, I included meta descriptions as part of the test. I think part of me inside didn’t think it really mattered, so eventually I stopped updating them as well as part it.

      I’m sure they could make a small impact, but not huge. My main strategy there was to get Google to bold the main KWs in a “triangle” which I’ve heard made a difference in PPC, but I stopped doing it after a while.

      I’m sure it can make a small difference but the effort on a per description level is much higher. Would love to hear of any reports back if you try doing this yourself.

      • Benjamin Beck

        Thanks for the quick response Ross! I would love to see a test for CRO of Title Vs Meta Description optimization.

        I made a list of ways to improve meta descriptions that can also be applied to title tags. http://benjaminbeck.com/serp-optimization-meta-descriptions-optimization/

        Thanks again for the great post. Very refreshing compared to all the “SEO 2015” posts in my feed.

  • Luissa Tallo

    Ross, Thanks so much for laying this out so clearly. I’m a content writer, so I don’t often think about the title tag as much as I should. With your help, I now see how important it is for the user to have these references in the titles that will grab their interest with specific info.

  • Cody

    What if we are not receiving over 2,000 impressions for our pages… Can we take smaller sample sizes and make it work?

  • Chris London

    Nice article Ross, I appreciate the examples and recommendations.

  • Cody

    Hey Ross,

    What do you think about smaller sample sizes… Most of our pages will not reach over 2,000 impressions in just 2 weeks…. Any thoughts?

    • rosshudgens

      Cody, if the pages are similar, such as the Moz profile page reference I made above, you can group them and them measure changes to traffic in Analytics. If they are all unique, you might try testing how people respond in your industry to a few higher traffic pages. If CTR goes up, that same practice will likely also help you on the lower trafficked pages.

  • jeypandian

    Where’s the LinkedIn share button? 🙁

    • rosshudgens

      Jey we did an analysis of which network was sending the least traffic and LinkedIn was at the bottom of the heap, so removed it due to space limitations. Sorry. 🙁

      • jeypandian

        That’s interesting – but what about traffic quality? I share only the best articles on my LinkedIn while the above average short attention span goes to Twitter. I’d imagine LinkedIn traffic is more engaging – thoughts?

        • rosshudgens

          Quality is worse, too. Since June, our average session duration from Linkedin is actually 40 whole seconds worse than Twitter. Only 6 seconds worse than Facebook, but also less traffic, too. Your mileage may vary, of course.

  • http://www.incarcamera.org/ Carcamera Secrooll

    It seems use number in title will attract more eyes

  • http://thomasoshaughnessy.com Tommy O’Shaughnessy

    Great article. I often get bogged down in the technical aspects of SEO, and you make a strong argument for focusing on the psychology of the user to increase CTR. Have you considered reaching out to the companies you mention with your article? Could land some new business.

  • disqus_1GDRttZXes

    Thanks a lot do you have an exemple for real estate ? That would be awesome 🙂

  • http://www.godaddy.com/ Jim Christian

    Great Post Ross! Love it.

  • Howell Glen

    What’s your thoughts on “review stars” schema showing up in the SERPs for some verticals? I’d bet it helps CTR.

  • JHTScherck

    This is *exactly* the inspiration i needed. Great read, smart strategy – thanks for sharing.

  • Graham Glynn

    Great article with excellent examples, this is powerful stuff!

  • http://thedigitalelevator.com/ Daniel Lofaso

    Ross, great post and solid insight into examples of PPC ad copy that plays to psychology that should be adopted by organic title tags for CTR. You’re absolutely right, many SEOs are still using “2005 title tags” and not considering how much smarter Google has gotten at allowing us to get away from over-optimized title tags with 9,000 keywords in them. I’m wondering if you have any data on local results that, according to correlation studies, rank better with the city/state in the title tag, thus taking up valuable character space?

  • http://www.branding.fm/ Joshua Cabe Johnson

    Hi Ross,

    Great clarity for the overused and heavily seasoned title enthusiasts!

    I slightly disagree that A/B split testing is impossible to determine if you are using ad platforms like Facebook to send traffic.

  • Dave Fogel

    I really enjoyed this post. Title tags are tricky and nobody seems to know the best way to do them anymore. I think you found it!

  • Mihai Aperghis

    Excellent stuff Ross!

    CTR is probably one of the most overlooked factor when it comes to on-page optimization, and it’s really awesome to see you delved thoroughly into this.

    I actually mentioned this article in the last Google Webmaster Office Hours hangout, where I talked to John about the balance between title relevance and user interests: http://youtu.be/SDcwFPdJrZc?t=1h3m45s

    Looking forward to more awesomeness 🙂

    • rosshudgens

      Thanks so much Mihai! I think that let John to share on Google Plus as well, which I noticed. Much appreciate you bringing it up and glad he supported it as well.

      • Mihai Aperghis

        My pleasure 🙂 I actually shared the link with John in the hangouts chat window. Please please keep writing more awesome stuff!

        PS: I wanna get your opinion on something, mind if I hit you up on your Gmail?

  • Miguel Salcido

    Great post and love the tips and examples at the end of the post.

    I too am a HUGE fan of GWT, especially ever since we lost keyword level data in Google Analytics!

    But I’m still pretty skeptical of A/B testing Title tags because there are just too many outside variables that can affect it and also because the data is not the greatest.

    Let me first address the data issue with GWT. You’re looking at the page level in the Top Content section. And the average ranking and CTR are based on an average across many keywords, in some instances hundreds of different keywords that get that page Impressions.

    Then there are the obvious issues with SERP CTR. With hundreds of different keywords getting different CTRs based on SERP position… well, it creates a bit of a data mess.

    Ultimately, I think that the idea that this work is only useful in looking for larger gains helps to negate all of the outside variables. So as you said, a 20% gain could easily be counted as a win. But looking at 1%-2% gains is probably doesn’t give a statistical significance.

    As a parting note/question, I’m imagining that this is only a good thing to do on sites that are matured and fairly stable in terms of organic traffic growth and authority. Because newer sites tend to have more erratic growth.

  • http://www.custard.co.uk/ Matt Fielding

    “With volume references, a best practice is to ping dev to auto-update your title tag with the newest number of said references.” I’d love to do this for a couple of e-commerce clients, do you know if this is possible in Magento?

  • http://www.nicolegrodesky.com/ Nicole Grodesky

    Great post! We just reworked all of our product level title tags. We looked at competitors and did research on pipes vs. commas. It was hard to find any recent topics on the subject because people stopped believing that title tags and metas don’t really matter. I talked to a CEO of an SEO agency that didn’t think title tags are important anymore. Let me tell you that after we restructured our title tags we’ve seen a pretty sizable in keyword rankings and keyword specific traffic generated to targeted landing pages. One question, what’s your take on pipes vs commas in the title tag? We went with pipes – just so you know.

  • http://www.digitalrial.com/ Ramesh Sanapala

    I will follow this golden tips to my pre-launch On-page for my website

  • http://www.smartwebsolutions.com.au/ SEO and Web Design Guy

    I needed this article, I have a number of sites that I now need to go back to refresh title tags..

  • http://www.infosolutionsgoa.com/ Joseph de Souza

    Interesting article. However in mobile searches larger title lengths are being displayed. Specifically I have counted up to 73 characters being displayed without being cutoff.

  • http://luizcentenaro.com/ Luiz Centenaro

    Epic post Ross, super helpful and practical. Testing new Title tags for my ecom shop HammockTown.com

  • http://webmechanix.com Will C

    This is incredible!