How’s that for a confused, or at least confusing article title?
I posted a blog article last week about some DIY stuff which wasn’t particularly noteworthy and truth be known I just wanted to post something to see if I could test a fix for the authorship tags on the site.
Back when authorship was just a toddler in the Google suite of obscure and not so obscure tags I went with some advice from somewhere to put a link with ‘rel=author’ on every blog post page to my profile page and slap a link on the profile page to my Google+ profile and I’d be done.
That worked for about, well. I’m not entirely sure it did. For exact match entire passages and phrases from my posts I’d sometimes seen my face staring back at me from the search results, but mostly nothing changed.
At work however we have a blog contributor who is consistently showing up as his miniature self smiling beside search results for his posts even though none of the requisite link tags are in place.
We have no links to his Google+ profile anywhere on the site and the only part of the authorship puzzle that’s been met is the contributor entry on his Google plus page.
I’m not going to go into any detail about how to make authorship work, there are a lot of good articles around the web on how that can be done and Google’s own help pages are as good as any now that it’s well established.
After the page was indexed fully I ran a range of different test searches which told me that authorship was working along with confirming a bunch of other odds and sods that should be common knowledge if you’re in the online marketing game.
What I found interesting though is how subtle search phrase changes changed when authorship shows up in the results or when it doesn’t. Equally I discovered some small words that made differences as well when I normally wouldn’t expect it.
So, without further delay, a pile of search results screenshots with comments for each…
￼First up we have a mixed up phrase from the blog post, and I’m top result. That’s mission 1 achieved, the page is indexed and we can move onto testing some other ideas out.
As a group of keywords ‘portable risks side note’ is not that stunning but you can see immediately how less than ethical SEO companies might convince a customer that a set of keywords are critical and get a rank for that combo under the guise of long-tail search. Followed quickly by the bill and a rapid exit to the nearest hills.
Long story which I can’t really post about, but I recently helped a friend with exactly that problem who’d paid handsomely for an SEO consultant to get their pages to rank well for a totally useless set of keywords.
This stuff is not rocket science but if you want to be top hit for ‘used car’ that is a whole other can of worms and requires a lot more effort as the content I’m using for these test searches is not really what happens in the real world.
An interesting thing to note about this search result is that the snippet of text is not the meta description for the page.
SEO tidbit #1 from this blog post: No matter how much time you spend crafting the description tag it may not show up in the serps these days if the search terms don’t match the description.
Oh, and the authorship worked. Who’s that attractive looking chap beside the search result?
I did a bit of messing about with combinations of keywords and found that this one still gave second place result but dropped my authorship. Again the search phrase itself is pretty meaningless but it highlights something about Authorship.
If Google doesn’t think who wrote the article is that important to the search results you wont get the extra credibility in the search results page. That means if you’re struggling with testing the markup pay a bit more attention to what you see in Google’s structured data testing tool and what you’re content is about rather than just trying to get your photo up on what you think the page should rank for.
Note that the snippet is different again. Still nothing from the description tag. Instead this time we have a mash-up from two paragraphs highlighting where the algorithm says the keywords were found within the body of the content.
A simple change here. Removed ‘on’ and there’s 70,000 or so more results found in the index but it doesn’t change the top few results. The fact is that small words sometimes don’t matter, despite how much your english teacher might have insisted otherwise.
Clearly if you were prepared to click a few more pages into the results you’d see a difference though, so let’s try something different.
Same words with the ‘on’ back in the mix with a different order and we’ve dropped a couple of hundred thousand potential results even though the top three results have not changed.
So, the order of small words does matter. It would seem that the combinations of ‘on side’, ‘on note’ and ‘note on side’ are probably more common in content than ‘on portable’.
I’m obviously mincing my words, almost literally, to make a point here.
When in the English language you write, order important it is. Unless you’re Yoda that is.
Google have long said that well crafted content is important and phrasing that is common to your target audience is going to rank better than the best writers missive or random words on a page that used to be common in the AltaVista days.
As a total aside, if you’re interested in SEO and don’t know what I mean by AltaVista days, you missed out on a golden age for SEO consultants that allowed people to do all sorts of things that would get them kicked from the index of even the slackest engine now. Ahhhh, those were the days.
Another shuffle of keywords and the third result has vanished down to about position six although cbsnews and I are still batting pretty well for some obscure text.
‘Notes on’ in this case is what starts the page title tag and the first H1 on the page for the result that’s popped up to number three on the hit list.
That right there is old-school SEO advice. Have relevant title tags and heading structures with text people will search for. If your page is about tomatoes having the page title ‘Shoe leather replacements for tomatoes’ and the first H1 tag the same will probably get you more search traffic for shoe leather than it will tomatoes.
One more shuffle of keywords and this time a more correctly constructed phrase from an English point of view and it’s got four of the five words in the same order as my post so the dashing fella on the left of the search makes a sudden re-appearance.
So even though this is not an exact match to the text the algorithm calculates that the order makes better sense and is more likely to be well structured content deserves that little bit of extra attention the authorship gives.
cbsnews.com is still there but lets face it… If my site had as much link juice as a major news site I’d have Google adsense on here and be counting my sports cars parked in the garage of my French Riviera holiday home not writing this for entertainment.
The osha.gov site appearing there is interesting, but again .gov sites have credibility oozing from their TLD so nothing surprises me when I see them showing up in search results.
Now for a little image searching using ‘testing FT-857’ seems like a pretty good image search term if you’re into amateur radio and want to find out about the FT-857.
The image is result four which is a good slot and your SEO handbook will tell you the image names are all important for such things and the alt tags. Don’t forget the alt tags.
In this case the alt tag is indeed ‘Testing on the FT-857’ and searching for exactly that will bring the image up to the top hit, not the lowly number four slot.
What about that image name? It’s actually ‘130818-171341-0001.jpg’.
Correct and contextual naming of images is a good idea but don’t forget the auxiliary tags around images. The only place FT-857 appeared before this post on my entire website is in the alt and title tags for that image.
Better than that, this search gets me top hit for a a combination of keywords from the page and FT-857 which only appears in the alt tag for the image and the title tag for the link to the popup copy of the image.
If I’d bothered to name the image in a useful fashion I could probably rank for some useful phrases as well as that one. This is basic stuff but day in day out I see SEO advice about all sorts of other things. Getting the basics right on this is going to get me traffic for people testing FT-857 Radios with power pole connectors.
One last screenshot to round out the observations for the evening. An image search for ‘gel FT-857’ showing a top hit for my photo. The word ‘gel’ is not in the alt tag for the image, but it is in the title attribute for the link to the popup.
If you hang plain english title tags on links to images and content you can improve their positioning for key words and phrases in the linked content or in this case can give you a ranking for a term that does not exist anywhere in the content apart from the tag.
By way of a disclaimer and for the sake of completeness: I did these searches from a New Zealand IP on www.google.co.nz, using google chrome in incognito mode to avoid search history slanting the results. Your results may vary if you’re in a different country of have substantial search history for similar terms or sites. Some of them were on my Ubuntu Desktop and the balance on a Windows 7 laptop, because I happen to be sitting in front of the telly pretending to watch something, so the fonts look slightly different in some of the screenshots.
(I did do a bit of testing from a US IP using google.com in incognito mode and got very similar results, although the serps were slightly different the observations would be the same. If you’re reading this more than a week after I wrote it the search results will probably have changed, the web is a dynamic place.)