Donna's Kooky Theory

[rmccarley]

http://www.seo-scoop.com/direct_link.cfm?thepost=546

DD noticed some changes in her SERPs for seasonal KWs. Does Google use popularity as an indicator?

[pops]

This sounds like the kind of fun thing Google might like to do. After all the relevance of certain cites does vary with the season. It would be interesting to test this somehow around other holidays. It seems like Google could take this into consideration without the popularity of the site.

[rmccarley]

I'm wondering though... Would it be a popularity thing like Alexa as she described, or could G alter their algo based on a calender? Popularity can be faked.

By knowing that there are seasonal interests and that people looking for information off-season are interested in something else.

For example, a search for [Mother's Day] in December is more likely for information *about* the topic (history, traditions, date of, etc.) than ways to appease (purchase flowers, etc.) the topic.

[pops]

I would think that a lot of seasonal sites are also only updated as the season approaches. That might cause a temporary spike in the ratings which would then drop off as the content ages and nothing new is added.

[skitzzo]

Originally Posted by skitzzos_pop

I would think that a lot of seasonal sites are also only updated as the season approaches. That might cause a temporary spike in the ratings which would then drop off as the content ages and nothing new is added.

Oooh... pulling in my theory of irregular content.... nice tie in pops... this is really interesting idea... it could have BIG implications for retail sites etc that depend on the holiday season sales for the bulk of their profit...

[Too Bad]

I can see a direct connection (and explenation for this) in this thread:
http://www.seorefugee.com/forums/goo...rate-algo.html

[skitzzo]

Originally Posted by Jammer

I can see a direct connection (and explenation for this) in this thread:
http://www.seorefugee.com/forums/goo...rate-algo.html

But if it were JUST based on Click through rate (CTR) then wouldnt it take a while for the content to shift to the top? Using RMC's example, it would take a while for the Gift information to move to the top for Mother's day. Google would be making a certain number of users jump through the hoops to find what they want to start the movement. I think this would be a much less accurate way to measure it rather than somehow incorporating the calendar.

Another point that hasnt been discussed yet is what about international holidays? I know that Mother's day here is not the same day as Mother's Day in the UK etc.

[rmccarley]

They have Mother's Day in the UK? Oh, it's only a difference of 2 days, so it would still fit in the same "season".

Honestly I think it's influenced by:

- A seasonal shift algo
- CTR
- Amount of searches, or search popularity (SP?)

Each one probably has some effect on the others. As to how much, only Google knows...

[dazzlindonna]

Time for me to chime in. So, throughout the year (this year), I very rarely even looked at my Christmas site. Then, in September, I gave it a quick once over, and did the following...
1. Added a half dozen images on one internal page (of a site that has several hundred pages).
2. Checked over my affiliate links to make sure they were still valid links. Removed one or two that were not.
Then, in late November, I made one other change - I changed the adsense ads from one size to another - on two pages (the home page and one interior page).

I did not actively seek any backlinks throughout the entire year. I also didn't really notice any new ones pop up, but I suppose there might have been a handful of unasked-for backlinks that came about during the year.

So, more or less, I did almost nothing to the site all year. Very, very minor changes is all.

From January to September, I had semi-decent rankings - mostly in the 20's range.

In September, bumped up to top 10's and stayed that way till today, when I dropped down to 90's and below.

Last year, when I dropped down, the rankings came back within a few weeks to the 20's. Will be interesting to see if the same pattern follows.

Yahoo and MSN - top 5 all year long - and still top 5.

Traffic: maybe 10 users per day average until the season started. Once October came along, bumped up to 1000/day. Then, 10,000/day in November and 20,000/day in December (averages). Back down to 1,000/day right after Christmas, and will probably be down to 10/day within the next month.

So, there's the facts for my site. None of my other holiday sites seem to suffer the same problem, but it may have something to do with the Christmas season. Perhaps Google does something special just for that season. Dunno.

[rmccarley]

Not sure if Cutts is eluding to anything here, but I pulled this from his Blog:

Originally Posted by Matt Cutts

Jagger wrapped up by about November 18th. I tried to impress on folks at Google that to the extent possible, you don’t want to make major shake-ups to the index during the holiday shopping season, so people should either launch their changes relatively early (e.g. like Jagger), or try to hold off until after the holiday season is winding down.

It's like he almost implies a change around the holidays, but not really.