|Giving Google what they want - increasing visitor "satisfaction"|
Google likes sites that their searchers like, presumably indicated by such metrics as pageviews/visitor, time on site, bounce rates etc.
So I think it is important to discuss what we can do to make our visitors explore more of our site.
For example Wikipedia's embedded links throughout each page, make it very easy to click on certain related subjects, thus increasing their pageviews.
What other factors have you noticed that increase your visitor interaction?
Who says Google like sites that give searchers what they want? If a searcher is looking to buy an item. They want to go to a page and buy it. Not go through several pages etc. To find a site that lists the item but doesn't stock the item is a waste of the searchers time. I can spend a whole lot of time on a site searching through a rat maze and never find what I want. I too can go to a site and hit one page and find exactly what I want. To assume that searchers come to a site through the homepage is not always correct. I know I have customers that spend hours at a time on the site. The site was designed in response to comments made by our customers. Google has hit us with Penguin yet customers still find us on page 2 because we actually do have the product. I don't think Google truly understands business. Some buyers are browsers that look but don't buy, other buyers go directly to what they want and buy. Some people look and bookmark and comeback months later. That information is hard to track. The good sites will still get sales maybe just not as much with Google as it is today. Then we get the comments on poor customer satisfaction from people buying online. That is caused by putting some of these poor companies at the top.
What gets me is here we are implementing all the security and PCI scans and specialized shipping features etc to accomodate customers with options and tracking features. And yet I see companies without secure shopping carts, no verification etc being ranked higher. Yet Google says we are suppose to ask which site makes the customers feel more secure. Google gives a good speech but doesn't seem to follow what they preach. So are we building sites for Google or for our customers? I don't think what Google is looking for is the same as what our customers are looking for.
Look at amazon though, if you want to go and buy a specific item then you are inundated with other stimuli of related items, reviews to read, book samples, next thing you know, you have spent 10 minutes there and visited 5 different pages.
They can inundate all they like. What takes time is finding the Advanced Search so you can tell them exactly what you're looking for. Grr. (Book by author x, with word y in the title-- is this approach so arcane that you will only offer it as a last resort? They've started hiding it in Google Books too.)
I've yet to understand how g### gets usable information from those searchers who open a bunch of likely hits in concurrent tabs and then read them one by one, closing as they go-- meaning that the time spent on a page will be a function of the results ordering itself, not of page quality. The only thing they can measure is the combined time spent looking at the complete batch-- and then only if all of them were so useless it's time for another search.
I have to remember it myself too. The page-opening time shown in logs may not have anything to do with what time the visitor first physically saw the page. Maybe they didn't really spend 15 minutes. It was 14 minutes on some other site from the same search before they even got to your tab.
I tend to agree with what you wrote.
Google is still struggling to monitor and compare page performance accurately. They may have the data, which is very "noisy" and makes it hard for them to get what they want.
Lets aim this thread a little more in the scientific direction, since nobody has any stats to support claims.
|I too can go to a site and hit one page and find exactly what I want. |
You aren't, and for that matter the majority of your visitors aren't, the ideal visitor in which to get a baseline reading about your website's pages. Keep in mind that Google is VERY good at understanding web pages and is always incorporating new signals(social), tweaking old signals(bounce rates etc) and coming up with parallel ideas to add on top of the signal stream(Panda, Penguin etc) and then there's the stuff that would drop your jaw if you knew about, there are a lot of web pages out there to process!
The internet is squishy - when you do something to your site in excess something else bulges outward according to a recent Bing employee interview. This suggests that after getting a base reading of your site there are expectations of that site, a range so to speak, for each of the variables.
I just wanted to add those two factors in so that "measuring" satisfaction isn't trivialized too much, it's complex and then some. In short satisfaction brings possible backlinks, mentions and other goodies that any site is expected to acquire over time, even the rate of these is projected as some have demonstrated negative results when getting too few, or too many, backlinks.
What can you do? A couple of ideas:
- dedicate 10% of your ad displays to your own site
By replacing 10% of the advertising you show with ads that deep link to important sections of your own site you should see a positive bump in statistics like bounce rate, time on site and exit rate. These can't hurt and the ads might even help some visitors. You are always your own sites biggest fan, why not advertise it!
- avoid conveying negative signals. Search algorithms are very good at sniffing out schemes and plots and shoddy practices so avoid even the semblance of shady and see what happens. Follow your favorite search engine's guidelines and pay particular attention to what they are telling you is spammy. Sure it might pay more to stuff a big ad under the title of your page but in the long run is it really paying you more? Or costing you traffic?
For good "time on page", visitor reviews seem to do well at keeping visitors on a particular page. And where there are a fair number of reviews these seem to encourage others to write reviews.
So do videos
@whatson I fail to see that you have given Google what they want at all, I don't think you understand what google want at all
I have no idea what Google wants anymore. Maybe whatson is correct. I just looked at who is above me and I spent some time on both sites (so whatson may be right). Site one was an ecommerce site that didnt even use an SSL in the checkout. That I am sure gave customers that vote of confidence Google preaches. The next site talked about sex acts and serial killers on what is supposed to be a G rated subject. But I did spend 5 minutes reading the whole thing. So maybe I am wrong and whatson is in a way correct. These sites did get me to spend my time there. Was a total waste though.
|I have no idea what Google wants anymore |
What google really really wants is your money!
You know I tried to give them that and even that didnt work. Odd too, they won't let me advertise on Adwords because they say it is a Trademark name (and they are correct). However two of my competitors advertise using Adwords using that same exact name. And the company that owns the name said they didn't give anyone permission. They tell me I can advertise outside of the country. I have too many foreign shipments as it is. Of course Google suggested I use a more generic term. I would get more business by just driving down the road throwing out dollar bills. Crazy part is one of the people advertising doesn't even have much of the product in stock. I really do not understand this whole thing anymore.
Hire someone who does and ask them if your expectations are realistic.
|I really do not understand this whole thing anymore |
|Giving Google what they want - increasing visitor "satisfaction" |
How do you know what Google want? Judging by ser pages, Google want you to advertise. They fill with ads, how is that user satisfaction or better for user? May be Wall Street satisfaction.
Google has said over and over again, quality content. Now I won't get into a debate what is quality content. In the last six months, I have built websites started ranking pretty quickly. The volume of traffic is increasing every month. The secret is quality content. We spend considerable effort researching and developing content for the website.I am sure Google takes notice of this. I don't know exactly how they measure it. But I know they can. They reward the websites with more traffic.
What if we forgot the assertion that anyone has any idea what Google really wants, and just focus on whatson's question: what can we do to make visitors explore our sites more? There's no doubt that would be a good thing to achieve, and even if it doesn't rocket you to #1 with Google, it should bring you more quality traffic (unless you've made your site entirely dependent on Google traffic).
Onto whatson's question: there's been speculation that google must just different niches by different standards. I.E., wikianswers should have be bouncing off after five seconds because it's a quick answer site. Of course, you also bounce off if you get to the page and no one's attempted to answer your question yet, so how Google can tell successful Wikianswer queries from unsuccessful ones, I don't know. Any thoughts?
Most websites however CAN be judged at least partially by time spent by visitors on site. What can we do to keep them around? Besides "writing good content" - I mean, we all know that, but it's like saying "To become a successful artist, you should paint better." I think we can get more specific than that.
+1 for diberry's first paragraph.
Assuming we want to increase people's time on site, stop them going back to serps and visit us again in future. Obviously a lot has already been covered on other threads
Here are some thoughts to get us going, some are probably a bit daft, some a bit obvious - but both the daft and obvious may spark someone else's imagination!:
Provide short user reviews - if it keeps people on the site longer than if they are just reading your technical spec
Get user reviews rated - if it makes reviewers want to come back and see how their review is received
Add a poll that people will want to revisit later to see the results - I guess polls that appeal to people's vanity are more likely to be succesful. I had two polls on my site - one was "why did you visit today" the other (which was relevant to the site) was "how fast have you ever driven" - the latter was far more succesful!
Say "bookmark this site and come back for next weeks poll"
Add video and promise a different video next week.
Give them what they want on the page they visit but offer a bit more on another page which is clearly linked to.
Tell them on the page that they are a valued visitor.
Nice design with clear navigation is more likely to keep people on the site.
For each specific page think "what would a typical visitor to this page want to read/see next"
Personalise your site so people feel they are dealing with real people, have pictures of staff even short biogs on the about us page, even pictures of staff on ordinary pages. (Have you seen Name.com's homepage?)
Thats enough from me for the mo.
|Google likes sites that their searchers like... |
I'll give you that one...
|...presumably indicated by such metrics as pageviews/visitor, time on site, bounce rates etc. |
I disagree with the above, because unless they changed something I've missed and started using Analytics for the SERPs they don't have access to the data for Time On Site or Page Views Per Visitor reliably enough and Bounce Rate has been stated as too noisy a signal to get much out of repeatedly, but they could use click-thru, click-back, re-click/re-search (a subset of bounce rate) a bit more reliably.
I think if you're looking too hard at the above areas you're probably missing something much more important.
A bounce is actually good in some situations (many, actually), unless you have an e-commerce site (a thank-you page visit would be a non-bounce), and even then it is if the visitor bookmarked the page and is going to buy from you when they get paid.
(Google does not have access to what someone bookmarks in their browser, unless it Chrome, but afaik, they are not using that type of info for SERPs, so it's a non-factor as far as rankings, and the only way they could possibly have direct visit data is if they get it from Chrome/toolbar, unless, again they have started using Analytics data for the SERPs.)
All bounce rate really tells you is either the visitor found exactly what they were looking for on the page or they didn't and there are a bunch of factors in your stats Google's Search Team (knowledge team, whatever) does not use (have access to) for SERPs which make the determination easier for you than it would ever be for an algorithm.
The timing of events may say more, but since a bounce happens any time there is a single page view to say a high bounce rate is bad has been proven false more than once. I had a site with a page that had a +90% bounce rate in Analytics, but the average time on the page was over 5 mins ... The page was #2 in the SERPs for multiple queries. I loved the high bounce rate, because it meant I did my job, Google did theirs, visitors found what they were looking for.
I would say trying to force visitors into more than one page view could actually be a [very] bad thing if it annoys other people like it does me, because you might get me once, but after you do I either block your site in the results or just refuse to click on yours, which harms your site way more than a single bounce because your page wasn't what I was looking for would. And trying to keep me there with the floating div I have to close works once too, but after that, I won't see it again, because you're either blocked or I just won't click ... My personal opinion is: When you try to manage the wrong metrics (or the right metrics the wrong way), especially when the management becomes detrimental to user experience, you do [way] more harm than good.
What type of 'quality signal' do you think it sends when I click on 4 sites, none of which have what I'm looking for, return from 3 very quickly, spend the most time on yours, and then block you from the results before my next query? Are you sure you want to make my (and other's) visits as long and as many page views as possible? Google can't 'see' what people did on your site after they left the results page, and even if they can, you can't see what visitors do once they go back to Google after someone sent them on a '4 page deep, close the div, click here, then there, then the other place' wild goose chase ... I'm sure there's more than one person out there who does the preceding on their site and has great looking stats, but for some odd and inexplicable reason their rankings have tanked ... I'll leave why to the imagination of the readers, maybe Google's just broken? The stats on sites that keep people there and make them view multiple pages look great, so if Google doesn't keep sending people it must be a problem with Google, right?
Increasing visitor stats is my main focus at the mo, and I don't think there's enough discussion about real actionable advice we can apply to our sites.
Three things that I've done that have shown a real boost are: adding more options at the end of articles, with back as well as forward options for article suggestions, and also better calls to action and social fluff.
Also, more breaking up of text blocks, especially in the first screen ATF, with quote call-outs and photos. Ironically this second change actually decreased stats a bit, but mostly because our traffic grew and that brought some less qualified traffic, but still substantially higher in absolute terms.
Finally, prettifying everything and really paying attention to smooth, simple and elegant design that puts the content front and centre.
I've noticed an interesting trend in a sector that I watch (finance). Many of the top-ranking sites have affiliate links on the main page, above the fold, with no navigation to inner pages. The internal pages have no navigation except for, perhaps, a link back to the homepage. In terms of site metrics, I don't see how visitors could possibly stay on these sites for long. Furthermore, how is it that Google doesn't see those internal pages as doorways (given that many aren't linked from within the site, I can only see them by doing a site: search).
If I search "best widgets", the #1 result on this extremely competitive financial phrase is the homepage of a website with 10 affiliate links above the fold, all "nofollow", target="blank". Not a single link to an inner page. I see several other sites with this same structure. Site metrics? The user comes in and clicks an affiliate link...that's it. The only factor that could possibly keep the visitor on the site longer is that the affiliate links open in a new browser window, but I doubt this has a big influence. What bothers me is the lack of usable site navigation, which many of the pandalized/penguinzed established sites had.