Back button and Panda

[johnblack]

This may be a bit of a nit-picker question but in recent Panda discussions and general SEO posts, we've discussed the fact that SE are now looking at how quickly the user hits the 'back' button when they click on a SERP entry.

However I've noticed that when I do that using Firefox 8.0, the SERPs page is cached and thus there's no re-load from the search engine's server. So how do they know I've hit the back button?

Is it more that they look at how quickly you click on another entry in the SERPs or go to the next page?

Sorry if this sounds a bit nit-picky but I'm just trying to get my head around the whole Panda thing (I realise this is just part of it) and what the search engines are doing or looking for it terms of user experience etc.

[tedster]

First let's say that the short click versus long click metric has been a QA check used by search engines for a lot longer than Panda. It was brought - by the Bing rep - at a conference this year where both Bing and Google people were on the stage. They both agreed that short clicks are one signal they watch.

I don't think we know in specific detail how a short click is measured. Some methods seems clear enough (Chrome users, toolbar data, etc). My assumption is that the search engines are content with sampled data, that and that they know the signal is going to be a bit fuzzy.

I also don't recall anyone one from Google saying, specifically, that short click is definitely one of the Panda signals. It certainly might be in the mix, but we don't know that for sure.

Clearly a short click is a query specific signal. If your site is getting a high bounce rate with low time on page for an important query phrase, that's something worth addressing. If the query is not a well targeted term, then Google "should" lower your URL for that term. In other words, don't chase after poorly targeted traffic just to "fix" a bad Panda hit. In fact, I might even try to "de-optimize" for a term like that,

[johnblack]

Appreciate the reply tedster.

Yeah, I guess the short click metric would have been around quite a while, relatively easy to measure to get at least a rough idea of what the user thought of your page. I just see the metric being discussed a lot in Panda threads on web master world and other forums around the net.

I guess Panda would have been more a 'content analysis' type of update.

[deadsea]

I have seen evidence that Google is looking at user satisfaction on a query and page level for about 3 or 4 years in ways that can drastically move a site up or down in the SERPs. I have an assumption that Google uses the data on a site wide level as one of many signals in Panda. Probably a powerful signal. But I don't have any clear evidence for that.

I don't know exactly what the metrics are or how they are measured by Google. There are probably certain browsers, plugins, and behaviors that are easier to measure than others. For example, I often open many sites in tabs in quick succession from one of my searches before I look at any of them. In that case, I doubt that Google can get much useful information out of me about which one I actually found helpful.

[tedster]

relatively easy to measure to get at least a rough idea of what the user thought of your page.

I'd take it just a bit further: they find out what the user thought of your page as an answer to that specific query. If you get a fast click every time your page appears in the SERP no matter what the search term, then that "might" be more of a Panda metric.