In brief, if you’re targeting a specific search phrase, you should certainly make your title tag as close to that phrase as possible.
(Ideally, the title of your content should be the same exact phrase. So if you’re going for “Salmon fishing tips”, your title tag should be “Salmon fishing tips”.)
In my keyword research, I have seen numerous examples of relatively weak pages getting top-10 placements in the search results because their title tags begin with – or contain – the target search phrase.
Clearly, title tag contents have a big impact on relevance and ranking.
But have you ever stopped to wonder why?
Why Do Titles Matter?
Obviously, on one level, the title tag tells search engines what the page is about. But we know that on-page factors are only a small part of the picture.
It’s the off-page factors – what we call your “inbound link profile” – that have the biggest impact, particularly in competitive sectors.
Well, your content title has a big influence on your inbound link profile.
Analysis of Inbound Link Text
I looked at several of my most highly-linked pages using OpenSiteExplorer.
I thought it would be interesting to see what the link text of those inbound links was.
Because my site has never had any deliberate link building work done on it, this should show me a natural pattern.
My hypothesis is that many of those links will use either the <title> tag text, or the page’s main heading text, as the link text.
And link text is INCREDIBLY important for relevance.
When you get new links, the link text will always be one of..
- The title tag
- The main heading (H1)
- The full URL of the page
- The title or domain name of the website
- Some other sensible description (which contains some or all keywords)
- Some other random text
So I think it is more a case of correlation than causation. Things like the domain name, path, and <title> tag may not be direct ranking factors, but do contribute indirectly to getting inbound links that contain the right keywords.
Highly Unscientific But Useful Results
I analysed 190 links that point to five of my top pages. Although it is not a big sample, this gives us a rough idea of a natural inbound link profile.
|Type of link||Number found||Rough percentage|
|Main heading (H1)||11||6%|
|Site URL or name||36||19%|
|Other (with keywords)||51||27%|
What This Means to You
39% of the links I looked at used their own link text. Most of that was relevant and useful. This is good, because it’s natural to have a spread of link text.
Good content will always generate a more natural profile that follows this kind of head/long-tail pattern.
I would draw the following lessons from this quick experiment…
- Choose your domain name carefully. It is likely to pick up link text from most of your content that gets linked to. Keywords in domain names count.
- That’s why exact-match domains have still been so strong for years… although Google’s Penguin update may be changing that. It is currently still too early to tell.