Title Tag Capitalization Impact on SERP Rankings and Click Through Rates
-
My company made a branding decision to use lowercase for all of our title tags. This, of course, means that our titles on SERPs are all lower case. Overwhelmingly, it seems that websites use title case. This makes me wonder if we're shooting ourselves in the foot.
Does using lower case titles negatively impact our rankings and/or click through rates? Is there any data out there suggesting that title case has a better click through rate than lower case?
Thanks for reading!
-
Thanks John!
Your company's optimization tools are very interesting. I'd like to implement if possible...Do you have a way to integrate it w/ Amazon, similar to SERP Turkey's "Mechanical Turk?". See: http://www.tomanthony.co.uk/blog/amazon-mechanical-turk-serp-turkey/
I suppose for a small business, you would recommend running an optimization tool like yours or SERP turkey, then tracking the implementation w/ before and after from WMT using spreadsheets, correct?
If so, what is the minimum amount of impressions that should be collected to consider the data reliable? Our industry has relatively low search volume, so I would venture to guess that a set minimum time frame would make less sense than impression totals.
-
I do not think anything can be truly accurate or perfect online, the nature of bots, broken links, and again too many variables. But it is 100% improvement on the old search console. In short it is unquestionably more accurate than the old search console. The old search console had too many variables. The new search console, gives us more succinct information.
And on the cross comparison of data between old search console and new search console - I think this would have negligible utility - certainly not worth relying on. The way the data has been collated means it is not like for like. So yes, I think your experiments would be thrown off, but general trends maybe still be identifiable.
-
Hey John,
Thanks for your response. Do you believe the "Search Analytics" data in WMT is accurate? I just finished an experiment where I tracked clicks, impressions, CTR and avg. position (using data from WMT). However, I know that WMT data accuracy was an issue in the past which Google has supposedly fixed.
Since I started the experiments in mid March before the data accuracy was addressed, do you think my test could be thrown off?
Thanks!
-
Hey TheDude -
For my rudimentary test I recorded 28 days worth of WMT data (impressions, clicks, CTR, avg pos.) before the capitalization change, then measured 28 days worth of data after the change and measured the difference.
My results were inconclusive, but in favor of title casing. My company's SEO traffic is seasonal, so clicks and CTR have been on the rise independent of the capitalization changes. This makes it difficult to isolate the impact of my changes.
If I were to do it again, I would identify N eligible for change pages with comparable stats and then divide them 50/50 into test and control groups. Then I'd make the changes to the test group and measure. The test group lift vs. control group would have given a clearer picture of the impact of the capitalization change.
-
The dude
Yes we get uplift. But for small traffic sites it can be close to immeasurable. For large traffic sites, we find it easy to get data from WMT's to demonstrate uplift. I have not heard from solid gold, except he seemed to like the experiment results back when we did it, and was implementing.
On us, yes, we use our own tool, but the results are independently measurable from WMT's. The new updates in search console earlier this year, have been integral and precise to tracking impact.
So suggest you excel spread sheet keypages CTR & Clicks (data from WMT's) and then monitor changes post implementation. You cannot go wrong for more than a week...
-
Hey Solid_Gold and John,
I'm in the middle of rewriting our title tags to increase organic CTR as well. Did either of you find conclusive results with your tests?
John, I'm assuming you used your own tool Predikkta for testing, correct?
Solid_Gold, I see that you said you'd also run your own before/after tests to determine which title tags had a higher CTR. How did you go about testing this?
Thanks guys!
-
SG
We do consistently identify that customers in that 1/3 of a second when they scan the page are attracted to click on the "well written" & "proper english" result. There are rare exceptions but does not look like they are relevant in your case. Good luck with it, at least I think you have partially answered the branding decision with some hard data...
-
Very cool, John. Thank you for running the test.
What I'm seeing in the results is that all of the title case title tags are projected to outperform the lowercase title tag I selected as the baseline. Granted, your software warns "Not enough responses & care should be taken when using the predictions", the outcome supports my hypothesis that title case would perform better than all lower case.
From here, I intend to run some before & after tests on our title tags to see if I can measure differences in performance.
Thanks again!
-
Ok. Your experiment is live:-
website www.predikkta.com
username: chasingfireflies
password: 2mvs7h8vLogin and review results... It is live now. Good to see what you learn.
-
That's correct!
-
No problem and the keyword phrase typed into google appears to be "girls halloween costumes". Correct me if I am wrong.
-
Great, John. Thank you.
Assumed URL: http://www.chasing-fireflies.com/costumes-dress-up/girls-costumes/
8 title tags:
I'm most interested in measuring the difference between all lower case and title case, so I've included 2 versions of 4 title tags:
- girls halloween costumes | wishcraft by chasing fireflies
- Girls Halloween Costumes | Wishcraft by Chasing Fireflies
- girls costumes | costume & dress-up | chasing fireflies
- Girls Costumes | Costume & Dress-Up | Chasing Fireflies
- girls costumes - girls halloween costumes
- Girls Costumes - Girls Halloween Costumes
- girls costumes for halloween | chasing fireflies
- Girls Costumes for Halloween | Chasing Fireflies
8 descriptions (full disclosure: I've taken these from competitors' pages on SERPs. I assume that your test does not involve publishing any this. Please let me know if I'm wrong and need to provide original descriptions.):
- Magical dress-up and Halloween costumes designed for her and her doll, from Chasing Fireflies.
- Your little girl deserves to look her best this Halloween. Find everything you need for girls costumes & kids Halloween accessories at Chasing Fireflies!
- Cute Halloween costumes for girls at affordable prices. Shop for Disney girls costumes, animal-themed girls Halloween costumes, and more.
- Chasing Fireflies has the best selection of girls' Halloween costumes. You can find costumes for girls so she can look adorable this Halloween.
- The best, most popular Halloween costumes for girls are available at Chasing Fireflies. Everything from adorable classics to popular movie characters.
- Chasing Fireflies has girls' Halloween costumes. You can find adorable costumes for girls so they look as sweet as candy this Halloween.
- Girls Halloween costumes at the very best prices. Shop for girls Frozen costumes, princess costumes, girls Halloween costumes and more.
- Find popular girls Halloween costumes at Chasing Fireflies. Shop a huge selection of girls Halloween costumes featuring their favorite characters.
Let me know if this suits your needs.
Cheers!
-
For interest more than anything. A freebie. No cost.
We will not do a competitor test, we start with a self test, hence offering a self test. It will give you insight into the preferred way to present your title tags, descriptions etc. So you can run the baseline (current result) result against 7 other titles and descriptions see the outcomes. Can test up to 8 URL's as well, but assume the URL is static. This way you can play around with the capitalization and present some hard data to the people above...
-
Can you help me understand what you're proposing John?
You'd like for me to provide you 8 lower case title tag & description pairs from our website and you will test them against their title case counterparts? How would you do this? Are you offering to sell me Predikkta services or are you offering me a trial of some variety?
-
SG
Your keen to get it right. I can run an experiment or test it for you and get you a definitive answer. Hopefully settle the discussion.
If you list 8 possible Title tags and 8 possible descriptions - (include the ones your using now so you have a baseline) then I will run an experiment for you. Let you what combination is the most clickable. Can do it Monday.
So you can find out Monday if capitalized or not in your instance...
-
Thanks again for taking the time to reply, John.
I'm picking up what you're laying down about there not being rules.
That you said Hire Experienced Professionals | Thumbtack looks more professional and therefore might earn a higher CTR supports my hypothesis. To your point - such a change would need to be tested. In our case, most of current title tags are well within the 512px width limit, so truncation is not a big concern should we decide to switch from lower case to title case.
-
SG
We assume CTR is a google ranking factor - as it just makes sense ie we know CTR is a ranking factor for Adwords so we assume it is a ranking factor for organic search results.
On lower case -v- upper case - It really depends on a case by case basis - there is no other way.
We do find that proper capitalization, english and a clean looking title have a positive impact on CTR. But it depends on each SERP page.
I will give you a few examples:-
Lets say your title was the one below
Thumbtack - Accomplish your personal projects
If the company changed to the below, they would increase their CTR for new customers. Note increased Capitalization and simply looks more professional.
Hire Experienced Professionals | Thumbtack
So this supports your hypothesis. However on the title below - Capitalization for each word has been used so based on the above as a general rule - it should have a better CTR. However it doesn't.
Cosmetics SFG Australia, Discount Make Up, Skin Care, H
Buy affordable makeup & beauty products | Cosmetics SFG
So by using the title starting "Buy affordable makeup.." - the CTR increased significantly. Why? To obtain this optimum Title, for SEO and clickability the company had to use lower case to come within the 512 pixels so the Ad did not truncate - and hence lose its customer friendliness.
I appreciate you are looking for rules but there are none. There are some general rules. ie at present from what we observe the pipe (|) has a higher CTR than the hyphen (-) - so we would recommend using the pipe. Also www. has a higher CTR than no www. on domain names.
But every instance needs to be carefully examined.
Finally say you are in the gambling space - our research shows - that when you want to gamble - it really does not matter what the Ad states - as long as it is passable, not off putting - it is all about position only. However for other SERP's ie "travel insurance" customers tend to scan the page in more detail. On those SERP's spend the time getting the balance between SEO & clickability right.
Not sure if I have assisted your argument to get capitalization, as indicated it all depends on your SERP that you are competing on.
-
Thanks Richard - I'm bookmarking the snippet optimizer!
-
Thanks for your response, John!
Re: rankings - I was thinking that if we were hurting CTR by using all lowercase that this in turn could harm our rankings since Moz's 2013 Search Engine Ranking Factors suggests that CTR is a ranking factor. Do you disagree with this?
-
We typically just use a headline format where the 1st letter of each word is capitalized with the exception of stop words (when used - but we of course, try to limit the use of these types of words in meta titles!)
I wouldn't have considered that the capitalization could impact rankings, however, I could see where it could have an effect on CTR.
I'm glad I stumbled upon this question to read some of the valuable comments made so far on this topic!
- Amiee
-
I've found using carefully placed all caps in certain parts of the description works great. Both the title and the description should have some type of cliff hanger or teaser at the end too. Take a look at the below example.
The SECRET to Smart META DATA is to Catch the EYE!!
Meta Data that has COMPELLING calls to ACTION & cleverly seeded BUZZ words achieves
FANTASTIC results for CTR. Great descriptions leave you hanging for...And I always use the mofoseo snippet optimizer so I know exactly whats its going to look like in the search results (provided Google doesn't change it)
-
Hi,
There is no direct effect, having in lower case (title) impact ranking but it can decrease the CTR at some extent. IMO you should neither use all upper case nor all lower case. Best way is keep first alphabet of every word in caps.
Hope this helps.
Thanks
-
I actually specialize in your question. In short there is no data either way. There are general rules.
The balance between clickability and SEO is integral to success for big companies defeating or mitigating page rank ie competing for big keywords like "car insurance" or "home loans". Small companies we find the problem is too granular - they go for title for SEO only. Small companies in fairness do not have time to research the balance.
The prime reason why no-one can give a definitive answer to your question is the title tag is generally only allowed to be 512 pixels. 1 pixel over and it truncates. So often you have to go lower case (optimize clickability) to fit within the 512 pixels or make lower case type adjustments to get to the optimum clickable headline. That said a general rule it is best to go with "proper pronunciation or spelling". It has a clear positive impact on clickability.
Our studies generally show that a Capitalized appropriate word in the right place has a higher impact on clickability.
However to be clear if this changes the title tag make up, ie it now truncates then it could be a disastrous decision for CTR. Finally and i will not go into detail here - it depends on your competitors title tags and descriptions on the relevant SERP.
So I appreciate you will hate the answer - but in short "it all depends".
Finally on rankings I have not heard that it would have any impact. So I would discount that. Though one of the myriad of factors for google on rankings would be spelling and "proper english". Though we deal with many companies where their branding is lower case, even capitalization second letter and it has not been an issue for them. So as indicated on rankings I would discount entirely but someone may have alternate views on same.
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Schema - Measuring Impact
Hi All,I've got a question about measuring the impact of schema. We implemented FAQ schema on a few pages such that the questions and answers written on the page are now displayed directly on the SERP.Google Search Console now provides both total page clicks and rich results clicks (under Search Appearance). To get total clicks on the page, do you simply add up the total page and rich results clicks? Or do the rich results clicks represent interactions with the accordion, rather than a click through to the page? And if both clicks are represented differently, does that mean that the page is only sometimes being rendered on the SERP as a rich result?Any guidance would be appreciated!Thanks.
Reporting & Analytics | | yaelslater0 -
How to track google auto search suggestion click?
Hello Guys, In google.co.uk when I search SEL and google gives me option of different different sites and when I click on any one site then that click tracking I need. I have attached the screenshot to understand easily. Is it possible to track such things or possible via server logs etc? TV99h
Reporting & Analytics | | micey1231 -
I have a WP site which uses categories to display the same content in several locations. Which items should get a canonical tag to avoid a ding for duplicate content?
So...I have a Knowledge Center and press room that pretty much use the same posts. So...technically the content looks like its on several pages because the post shows up on the Category listing page. Do I add a Canonical tag to each individual post...so that it is the only one that is counted? Also...I have a LONG disclaimer that goes at the bottom of most of the posts. would this count as duplicate content? Is there a way to markup a single paragraph to tell the spiders not to crawl it?
Reporting & Analytics | | LindsayiHart0 -
Bounce Rate: what is it EXACTLY?
Hi everyone: we all know the term 'Bounce Rate'. I'd like to think i have a good idea of what BR is....but some things are not really clear to me. Time to call in the experts. Question #1: What EXACTLY will stop Google from considering the visit as a bounce? As discussed not too long ago in this topic https://moz.com/community/q/will-this-fix-my-bounce-rate
Reporting & Analytics | | BasKierkels
Ruben wrote: "..what it basically means is that someone clicks on your SERP, and then clicks back to google? But, it doesn't matter if they spent 10 minutes on your page or 10 seconds" Jessica Conflitti wrote a reply in which she basically said that it might be a good idea to have visitors click to a different page OR a PDF-file. That's where my confusion has been for some time now: Clicking on a PDF-document, an image in the page that opens with Fancybox, a link to a different domain? Or can it only be a different URL on the same domain? The way i would expect it to be:
Pages contain the GA-tracking code. So am i right by thinking that Google needs to have the same GA-tracking code to be loaded twice? Because only at that point will they have two datapoints. And only then will they be able to tell that the visitor hasn't left. By clicking a PDF-document - as described by Jessica - you wouldn't load the GA-code twice. So I would expect that clicking a PDF does not make a difference for the BR. Don't get me wrong: i like the article but it is this detail that throws me off. IF Google can read or capture these clicks, what other elements can be used to reduce bounce rate? Clicking on a YouTube-video embedded in the page? I'm asking this because i want to get this right. Question #2: how much weight does BR have on Time on Page, Engagement, etc? We know Google is taking a lot of things into consideration when calculating the value of a URL or domain. So how much should we care for BR if we know the Time on Page is good and a large percentage of people are frequently returning? How about your experiences or knowledge on that? Really looking forward to your replies and help on clearing this topic for me. And perhaps some other readers as well! Bas0 -
Should I use sessions or unique visitors to work out my ecommerce conversion rate?
Hi all First question here but I've been lingering in the shadows for a while. As part of my companies digital marketing plan for the next financial year we are looking at benchmarking against certain KPIs. At the moment I simply report our conversion rate as Google Analytics displays it. I was incorrectly under the impression that it was reported as unique visits / total orders but I've now realised it's sessions / total orders. At my company we have quite a few repeat purchasers. So, is it best that we stick to the sessions / total orders conversion rate? My understanding is multiple sessions from the same visitor would all count towards this conversion rate and because we have repeat purchasers these wouldn't be captured under the unique visits / total orders method? It's almost as if every session we would have to consider that we have an opportunity to convert. The flip side of this is that on some of our higher margin products customers may visit multiple times before making a purchase. I should probably add that I'll be benchmarking data based on averages from the 1st April - 31st of March which is a financial year in the UK. The other KPI we will be benchmarking against is visitors. Should we change this to sessions if we will be benchmarking conversion rate using the sessions formula? This could help with continuity and could also help to reveal whether our planned content marketing efforts are engaging users. I hope this makes sense and thanks for reading and offering advice in advance. Joe
Reporting & Analytics | | joe-ainswoth1 -
Someone mentioned us on facebook and brought 10k clicks to homepage.. how do I find the content?
So this morning our website started to go crazy with hits all coming from Facebook, and I am trying to figure out how I can see who was talking about us? All the traffic landed on one page specifically. When I look in analytics it just shows social, and then I can drill down to see they are coming from facebook, but that's it.
Reporting & Analytics | | DemiGR0 -
Another high bounce rate
Hi there, One of my top landing pages has an 81% bounce rate. http://www.snowbusiness.com/what-we-do/film,-tv-and-advertising.aspx My first thoughts are purely bad IA and usability, but i know there must be loads of other things people might identify. Thanks in advance, Ben
Reporting & Analytics | | SnowFX0 -
Low bounce rate; need help troubleshooting code
I've had an outside developer do a bunch of custom work in Google Analytics to get my site to integrate with Foxycart and accurately report sales in the ecommerce section. With a foxycart upgrade came more GA tweaking, and now my bounce rate is at 1-2%. I know this isn't right, and suspecting there is something triggering GA a second time, causing a second page load, or something. Could someone that loves code look at http://www.strikemodels.com/ and tell me if they can easily spot what's obliterating the accuracy of my bounce rate calculations? Or do I need to go back to my dev and up the can of GA worms to troubleshoot things? As you can tell by the code, I'm running the latest version of WP with a few plugins, Thesis 1.8, and on Apache.
Reporting & Analytics | | KeriMorgret0