Different organic rankings on different computers?

seachto7

Howdee,

Confused a wee bit with the following. I am seeing different organic rankings on different computers for the same search queries.

Is it down to location settings? Browser history? For example, the query might rank 3 on one computer and then differ on another.

I changed locations and got the same results. Are organic results influenced organically within a country? (Local search results for example).

I have been using 3 different machines in Limerick, and getting different results on each. Subtle in some, significant in others.

Anyone know if there is an "official" explanation for this. i.e. "hear it from the horses mouth"...

Cheers

TsuDhoNimh

seachto7 wrote: »

Is it down to location settings? Browser history? For example, the query might rank 3 on one computer and then differ on another.

Results get adjusted based on both localisation (either IP address or Google account settings if signed in) and personalisation (various factors including browser history, cookies, google account history, etc.).

You can disable personalisation easy enough, either through account settings or various tools (like RedFly's Google Global - haven't used it personally in a long time, but I'll assume it still works a charm), but localisation now forms such an integral part of search quality for Google that even though it can be 'played with' it's almost irrelevant doing so in the context you're describing (information is always great, but unless it's actionable what's the point).

The users you're trying to target won't be taking the steps to disable these features, so going to great lengths to disable something like localisation is basically a fruitless task. It's the main reason that traditional 'rank reports' are now old hand and providing, at best, information with a huge caveat around it.

It's potentially far more useful to work with accurate data provided via WMT or analytics which takes both personalisation and localisation into account and provide information on where the result actually ranked when clicked (even that's of debatable value, given a result placed on the first page can have a 'ranking' of 14 with universal results, but that's a different debate).

seachto7 wrote: »

Anyone know if there is an "official" explanation for this. i.e. "hear it from the horses mouth"...

If you search the Google help documents you'll find plenty of official information on both personalisation and localisation (this type of thing).

Sometimes it doesn't quite give the 'full story', so no harm reading up some of the more authoritative 3rd party discussions relating to it too, seeing the results of testing and experience of others in the area.

seachto7

Thanks for that info. On a side note/ related note, I ran two reports using "rank tracker", and one was results compared to 1 month ago, the other was results compared to 3 months ago. Current rankings differed slightly. I ran the first one on Thursday, and the second one today. Surely, rank positions wouldn't change that much in 3 days..

TsuDhoNimh

seachto7 wrote: »

I ran the first one on Thursday, and the second one today. Surely, rank positions wouldn't change that much in 3 days..

Impossible to comment on 'in general'. That one will really depend on the specifics.

A site can go from 1st place to deindexed in the click of a button should a manual webspam penalty be applied. Other sites could easily slip from 3rd to 9th should some previously beneficial links suddenly be devalued for any reason. So even though they didn't trip the algo themselves, they suffer from others who have.

If there was an algo update that could have hit them, obviously that could cause large scale changes across the board. And even aside from the changes that occur on Googles side, something as simple as blog post changing location in site architecture (so being pushed off the home page by newer posts for example) can have enough of an impact on the link index to cause a variety of moves.

There are just too many factors at play, many of which have no clear 'timing', to give an answer to it in general. Well, the answer technically is "Yes. It can", but you see where I'm going with it.

******************

How does the 'Rank Tracker' you're using (I'm assuming that's the old browser add on one? SEOBook I think?) grab the data?

Do they do it completely independently of the user or does the machine (so browser/IP/Google account/etc.) have any bearing on it? If it's the latter, it could be part of the reason for the changes (but if it is the SEOBook one, I'd doubt if that's the case - though again haven't tested that tool in a long long time and can't for the life of me remember how it used to work).

seachto7

I'm using the "SEO Powersuite" Rank Reporting tool (www.link-assistant.com). I am only using it on the one machine however.

The different search results were achieved manually, just searching on different machines...

bceltic

This might help: http://searchengineland.com/guide/seo/personalization-search-engine-rankings

seachto7

Thanks

TsuDhoNimh

I'd do a little digging on the tool you're using and become aware of the exact implications of its use. It's not the end of the world if it is applying certain filters (wanted or unwanted) to results, but being aware of what it is or isn't doing is important when it comes to interpreting the results.

seachto7 wrote: »

The different search results were achieved manually, just searching on different machines...

Changing browser is enough to have a dramatic difference on the personalisation of results (especially so in the case where you've done the majority of client work on a single machine/browser/account which would build considerable links between you and that site), so changing machines could obviously have an impact there.

Probably no harm for you to run a few tests (more for illustration and personal experience than any need to 'track' the specific results) using various different browsers on the same machine (with/without browser history etc.), SERPs while logged into a google account vs. non logged in, disabling all personalisation vs. not touching it, etc. so you can see how much they can differ.

(Assuming you have a blog about SEO related topics, it's a decent topic of conversation if you do track the results. It's not exactly 'cutting edge', but getting info on it from an Irish context might get you some positive attention)

seachto7

I emailed the tool provider with my query, and they suggested tweaking some of the settings, so I'm in the middle of that at the moment.

Thanks for the tip TsuDhoHimh. It's certainly a curve ball that potential clients could well ask about. In my limited SEO experience, I have not experienced anyone asking that question in an Irish context just yet.

TsuDhoNimh

seachto7 wrote: »

In my limited SEO experience, I have not experienced anyone asking that question in an Irish context just yet.

Very few clients ever would, but it can happen (and has).

I know when I first got hit with it, specifically in an Irish context, it was a bit of a head scratcher.

I could speak in relation to factors involved and the impact it had in terms of international search in detail, but from an Irish specific point of view (so localisation variances between Dublin city, Dublin County and Meath for example) I wasn't so sure.

After plenty of testing I had plenty of firm answers and still lots of unknowns and head scratchers (although they were more in relation to some of the quirks of Irish IP addresses resulting from our irritating, at times, infrastructure), but it left me feeling a lot more confident the next time it was asked.

It's something that does change and has become more nuanced over time (with still quite a bit to go before it'll be 'right'), so just getting to grips with where it stands 'now' doesn't mean you won't have to look at it again in the future.