Title optimization, for USER psychology

Almost posted this as a reply to this thread since its the same logic. But, this really is an entirely different thing.

Another thing I've been thinking about that I don't see discussed...

Titles. Not optimizing them for whatever Google wants to see, I want to see some data/info on picking titles that get more clicks. If you're not an "i'm feelin lucky" person, it is safe to assume you're going to compare titles.

I'm going to guess you're either

1. In reality, looking for a long tail keyword and will click the title or description that best matches your search
2. Storefront shopping for the page that will give you want you want without needing to go back to the SERPs

For example, the query results for "best dog breeds" is:

1. Top 10 Best Dog Breeds of All Time
2. Quick Rankings: The Best Dog Breed, for You
3. Ultimate List of Dog Breeds
4. Dog Breeds 101: How to pick the best dog breed + list of top pups

Is it better to 1-up your competition (#1)?
Or offer more explicit value than your competition (#2)?
Or offer more value detail than your competition (#3)?
Do what #4 is already doing, and compile competitor's value into 1?
Click-bait it?


What type of approach do you take?

For obvious reasons, testing this in Google isn't something you can just do... but it could be tested in a study, and I'm sure the top-level seos think about this stuff.

There are Wordpress plugins that let you split test titles, such as this one: https://wordpress.org/plugins/wp-experiments-free/ (not specifically recommending it, it's just one that came up as I hunted for an example)

I'm not sure how, as I've not done it, but I suspect you could keep a steady meta title for Google as you split tested titles on your site itself. And then you could roll out the winner to the meta title as well.

It's tricky because a Title written for Google is going to help you rank, but CTR after you rank is going to help you keep the ranking and rank higher.

I don't attempt to do one or the other in isolation. Usually my posts are the most all-encompassing and that's what I want people to know, especially on these bigger keywords. So saying that while mentioning the parent keyword is my usual tactic.

Nat said:

Almost posted this as a reply to this thread since its the same logic. But, this really is an entirely different thing.

Another thing I've been thinking about that I don't see discussed...

Titles. Not optimizing them for whatever Google wants to see, I want to see some data/info on picking titles that get more clicks. If you're not an "i'm feelin lucky" person, it is safe to assume you're going to compare titles.

I'm going to guess you're either

1. In reality, looking for a long tail keyword and will click the title or description that best matches your search
2. Storefront shopping for the page that will give you want you want without needing to go back to the SERPs

For example, the query results for "best dog breeds" is:

1. Top 10 Best Dog Breeds of All Time
2. Quick Rankings: The Best Dog Breed, for You
3. Ultimate List of Dog Breeds
4. Dog Breeds 101: How to pick the best dog breed + list of top pups

Is it better to 1-up your competition (#1)?

5. Top 25 Best Dog Breeds of All Time --> More options than 1. and it still implies its ranked, so I can stop browsing after 15 if I want.​

Or offer more explicit value than your competition (#2)?

5. Quick Quiz: Find the best dog breed for YOU --> Sounds just as fast and possibly more personalized​

Or offer more value detail than your competition (#3)?

5. The Internet's Largest Ultimate List of Dog Breeds​

Do what #4 is already doing, and compile competitor's value into 1?

5. Ultimate list of 50 Best Dog Breeds + Quick Personalized Quiz​

Click-bait it?

5. Here's Why Petco Doesn't Want You Knowing the Best Dog Breeds​

What type of approach do you take?

For obvious reasons, testing this in Google isn't something you can just do... but it could be tested in a study, and I'm sure the top-level seos think about this stuff.

Click to expand...

Here is something I saw from TheBASystem about titles. It's more directed for social media posts but I use this as a guide for writing titles for my pages. I think it actually accomplishes appealing to both users and G.

3 Categories of Headlines: 1) Social Proof 2) Threat 3) Gain

Social Proof

• How [impressive number] Got [desired result] In [time period]
• Example: How 15 people sold their house in less than 45 days
• Example: How I learned to speak Spanish in 30 days
• How To [desired result] Like [world class example]
• Example: How To Bike Like Lance Armstrong (Use Local Person)
• Why I [blank] (And Maybe You Should To)

Threat

• Don't Try [blank] Without [desired action] First
• Example: Don't try that 30 diet plan without trying XYZ gym first
• The Ugly Truth About [blank]
• Example: The ugly truth about attorneys (if you're an attorney)
• Your [blank] Doesn't Want You To Read This [blank]

Gain

• Where [desired result] Is And How To Get It
• Example: Where the fastest cars are and where to get them
• You Don't Have To Be [something challenging] To Be [desired result]
• Example: You don't have to be the fastest cyclist to be fit
• How To Build A [blank] You Can Be Proud Of

They start talking about this subject around minute 31:00.

Basically, this can be boiled down to Internal vs. External concerns. "SEO" vs. UX. Or whatever. What I would say is, it's real easy for people to jump from one "bandwagon" to another, and sometimes something is lost in the transition. I would say the optimal solution probably lies somewhere in between. Clickbait titles, stuff that may generate good CTR on certain social platforms, etc. may do alright on those platforms, however they may be really poorly optimized from a technical standpoint to the point where they don't rank worth a damn.

Internal

• UX & CTR may be primary concern
• Consider in your site's code, using a custom title field for generating internal linking, that is separate from the actual page title
• For example, you could split test the custom field for page titles in a popular/related posts widget, to see if messaging changes effect a positive CTR improvement on certain parts of your site, reduce bounce, etc.

External

• "SEO" might be the primary, though UX and CTR are still factors
• Might consider starting on one end or the other (seo vs ux), let it roll, then try to find a happy medium between messaging vs technical optimization
• Probably best not to continuously change actual page titles and meta descriptions, on the off chance it may negatively affect stability of rankings (who knows, you'll need to test for your niche)

To explain it further, it's a bit like syndicating posts across multiple platforms. The average marketer, I would say, probably uses the same or very similar messaging across all of their platforms, and often probably doesn't see great results. You really have to determine, who is the audience, what are they looking for, and how do they want it presented. Same with your titles. Don't just use the same thing indiscriminately everywhere, figure out what's most appropriate for the audience, platform, and any technical concerns.

Just my recommendation but, I would probably start with split testing with custom fields for post feed titles and such, while maintaining your "SEO'd" actual page title. Save the actual page title changes for later, once you've found 1 or 2 variants that seem to be very effective internally. The general idea being, there are too many ranking factors to consider. If you are changing things frequently, good luck trying to decipher what may have had a positive effect on rankings.

I'm thinking only about user. Main KW is included, or some close variation and that's all to be honest. Also, I like to ask user a question in the title. For Google I have only KW or two.

Similar to what @turbin3 wrote - you have to figure out a way to balance UX and SEO.

What I like to do is use Yoast to create my title tag that will appear in the SERPs. Since Google is telling me that those "Top 25 Ways..." style titles are working (because that's what's ranking in the top 3 spots), that's the type of title I'll use within my <title> tags.

But, once I have the click on the page, I need to figure out how to keep them around. So, the headline in the <h1> tags will be something different than what I used in the title tags. This headline is all about engagement. I'll use it to speak to the reader directly in a conversational manner.

Between the heading 1 headline + my intro hook, I'm going UX 100%. If I can grab them with those two elements, I know I'll have a better chance at keeping them on site and potentially converting for some given action.

Great points all of the above. I've been doing research in CTR as well. Read a lot of articles and then tested in on one of my aff sites. So the conclusions from theory and testing:
- long titles (7-8 words)
- contain hyphens or columns
- have a question mark
- numbers
- brackets

all have better CTR. If you combine some of those in a compelling title, results accumulate.

I wanted to add some additional clarification, as I feel I left a few things out.

The style that works for search CTR is going to be different from niche to niche. Often, it will even vary significantly within the niche and individual SERP. You might also find that what appears to be working in search doesn't seem all that optimized in a traditional SEO sense, but actually seems a bit more UX-oriented. Imagine that! Considering that, here's a couple of simple things that might help bridge the gap between pure SEO vs. UX:

UX
Have you ever been asked a question that stopped you dead in your tracks? That's basically what you're trying to achieve.

It's often brief, sometimes subtle, sometimes blatant and/or outrageous, but it all accomplishes the same thing. You're asking them the question they know they need to hear, even if they weren't consciously aware of it before. You're telling them the thing they know they need to be telling themselves.

I would look at existing SERPs in the niche, or for closely related keywords, and determine if a good percentage of ranking pages are already showing a bias towards UX-oriented titles vs. bland, optimized ones. If so, at least match them and just make sure to incorporate a little bit of the following.

SEO
Here's one quick and easy way to make sure you're not missing out. Search one or several keywords related to what you're trying to optimize. Take note of everything that's bold in the SERP. Those are related, and some of which you might want to ensure you include in your title, description, and maybe the URL.

Does everything on page 1 have 3 or 4 bold keywords or phrases? Maybe more? Just make sure you're in the ballpark and incorporating some of those.

If you're on Chrome, a quick way is the Scraper plugin. When you right click on the SERP and select 'scrape similar', just use these XPaths to scrape the list of related keywords if you want to do this in bulk:

Google:
//em

Bing:
//p/strong

The point I'm trying to make is this. A lot of people still get caught up in worrying about certain technical and traditional details from the traditional standpoint of taking some general "best practice" and applying-to-all-the-things. Things like front vs. rear branding, front-biased keyword placement versus elsewhere within the meta element. These days, everything is highly variable down to the individual SERP.

For example, the one-size-fits-all of, "ALWAYS bias keywords to the front" is not necessarily always the best way to go. Sometimes, for some SERPs, it just doesn't matter a whole lot and the primary factors of importance may be other aspects.