Which one of these two helps you rank better?

Selected answers from the Dumb SEO Questions Facebook & G+ community.

• 
  Stockbridge Truslow: IMO, neither of those are useful choices. When a visitor lands on your site (from any source), the most important thing is that they convert to your goal. Neither "staying on your page" nor "clicking on another page" are your actual goals (though the click may be part of it). Every web page should be driving the visitor toward one of your goals - be it contacting you, qualifying them as a lead, calling you, purchasing something, or maybe even getting them to visit your store in the real world. I suppose there are exceptions to this - like on an ad sponsored news site or something. But in most cases, long dwells or arbitrary clicks are indicators of how someone might be moving along the path to the goal, but they are not goals in and of themselves.
• 
  Tony McCreath: I think it`s also worth thinking beyond that. If they triggered your goal 3 weeks later, that search click should take some credit as a touch point in the conversion path.
• 
  Stockbridge Truslow: True. With big ticket and complex items - the path can be long an circuitous. And there are other ways to be a step on the path (and altogether different goals than I`ve mentioned). Space and time prohibit an exhaustive description and explanation here. My point is that having to choose one of the above choice is generally useless without a whole bunch of other qualifying factors involved.
• 
  Tim Capper: So it depends on what the site is all about.If they landed on the Pink Fluffy Elephant product category page, you would expect them to scroll down the product page and then click through to a product.However if its the contact page or booking page, you would expect them to land and leave from that page.It all depends on what the search query is and what they expected to find on that page and desired action.
• 
  Chris Edwards: What is best is dependant on search intent. I have sites where my bounce rate is 100%, and my time on site less than 2 minutes. one site is making £120 each day every day (average)
• 
  Rob Woods: Neither is objectively more important. It depends on the nature of the query and whether the user got their answer from a single page on your site or not. Google actually has no idea whether a user goes to another page on your site or not. Simply bouncing back to Google also isn`t a sign of good or poor quality. If a user clicked on a page, got their answer, and went back to Google to search something else Google may infer that their question was answered on your page is a very good result for the query. What certainly can affect a ranking is if somebody goes from a Google search result page to your site, then goes back to Google and repeats the same query. That can very much be a signal that they didn`t get an answer to their question from your page, and if that happens a lot Google might infer that your page doesn`t deserve to rank highly for that query.
• 
  Rob Woods: Again, Google, for organic search purposes, has no idea how long the visitor stays on your page or whether they visit other pages on your site. Neither of those are going to affect your organic ranking.
• 
  Scott Hendison: Rob Woods Surely you didn`t mean they "have no idea" since between Chrome and Analytics they know exactly those things. As far as whether they apply that knowledge as an actual ranking factor, sure they`ve denied it, but it wouldn`t surprise me in the least if engagement was taken into account.
• 
  Rob Woods: Scott Hendison they have stated pretty categorically that they don`t use the data on the organic search side and while I think they can be vague at times and omit facts in what they say, I don`t think they outright bald-faced lie.
• 
  Scott Hendison: Rob Woods Sure, but they still ARE fully aware of the user engagement, so there`s nothing to stop them from using it as a ranking factor if they decided to. I think user activity can be a pretty relevant signal that could someday be taken into account with ease, don`t you think? I`m not talking about just time spent on site, but true engagement and interaction.
• 
  Rob Woods: Scott Hendison depends on the query. Some don’t require engagement to be a good result.
• 
  David Ogletree: If you have a phone number it is best that clicking or calling the number creates an event and registers an action so a bounce is not registered. That makes your bounce stats more useful. Also if you have long pages you can have a scroll register as an action so that a bounce is not registered.
• 
  Michael Stricker: Google likes actionable pages, where some kind of conclusive activity takes place (YMOYL pages). So, spending time on such a page that can be rated to exhibit EAT is a user reinforcing what Google already published in its Raters Guidelines. Any convertibility there is a good sign of quality for Google and hopefully has a good resulting business effect. If the landing page is not convertible, then the second ought to be. If neither are, then your win is in the form of in-depth content, perhaps scholarly, educational material... as long as the subsequent page is relevant and entity-related to the first, then topic and keyword reinforcement should occur, and better rank may result. So, short answer: “it depends.”
• 
  Marvin Hernandez: Thank you everyone for your input. I edited my original post because I probably could have done a better job clarifying. I just wanted to narrow it down to two choices…”Bounce Rate” vs. “Time on Site” although in this case the time on site would consist on being on one page only. So, if you think Google uses any of these two factors (among many others of course) for “possibly” pushing you up the SERP then which one do you think has more weight between these two only? Example: Visitor lands on your site watches a video only (about 3 to 5 minutes worth) then leaves as suppose to a visitor lands on your site doesn’t watch the video but visits another page on your site.
• 
  Stockbridge Truslow: Watching a video counts as engagement. There`s a click there.
• 
  Chris Edwards: Bad bounce rate would be Googles worse nightmare as it means they are sending traffic to a site that is clearly not relevant, as users are rapidly bouncing back to their search page.
• 
  Rob Woods: Unless users are quickly getting an answer and bouncing back to do a different search. Then it means the site is clearly relevant and efficient at satisfying the user’s need. That’s the problem with bounce rate and time on site. Neither tells google whether the need was satisfied or not and it’s why they don’t use it.
• 
  Chris Edwards: Well obviously if they do a different search :)
• 
  Rob Woods: Right. So it’s not bounce rate, it’s repeated searches for the same query. The behavior is on Google’s site, not the “target” site.
• 
  Chris Edwards: I disagree. My point was simple, if the searcher visits your site, bounces back then continues to click on other sites in the same SERP, then if i were writing the algorithm i would see that as a CLEAR signal that the site is not relevant to that search term. CAVEAT, I would also include data on how many other pages had the same result (fast bounce), and factor that in to the relevance.,
• 
  Ammon Johns: Bounce rates are a notoriously messy and noisy signal for any use at all. I remember one paper that said to use it at all they had to discount any bounce in less than 15 seconds as potentially just a misclick, and any bounce longer than 40 seconds as quite likely searching for a second opinion. That leaves a pretty narrow window for bounces suitable for any use at all, and even those are treated as a very low quality signal individually.It`s for this reason that I am certain that Google do not use bounce rates in the way most assume. They use them only aggregated, in mass, across all SERPs of a certain type to check what difference an entire algorithm tweak may have. And even then, it`s more about whether there are more or less bounces across all sites in a SERP, and whether a user needed to make more or less total clicks, on average, to reach a satisfactory result.Now, with machine learning there could always be outliers - cases where if one site gets a ridiculous amount of bounces (I`m talking above 90%) then perhaps that site needs to be looked at, and perhaps it goes to review. But in the main, bounces are used to determine if the algo is good, not if the site is. High bounce rates mean the SERP is bad, and the algorithm is promoting the wrong results.
• 
  Dan Thies:
• 
  Ammon Johns: To answer the question of which one would help rankings more the answer is "Neither".Google do not use analytics data in any way to influence rankings. I`m certain of that not only because so many google engineers have outright said so - in a company that is quite rightly rather paranoid of making false claims they could face legal issues over - but also because it could be considered as `unfair` monopolisation of analytics.Now, Google can use other data sources to get user data, and they do - mostly Chrome - but that`s not how they use that data.