|Statmarket Survey: Net surfing a dying sport|
64 percent of users arrive via direct navigation
Perhaps domain names are increasing in importance despite opinions to the contrary.
|...the majority of Internet sites worldwide are reached through direct navigation - typing a URL in their browser address bar or using a bookmark - rather than through search engines and Web links. As of February 3, 2003, over 64 percent of Internet users arrived at sites by direct navigation, compared to about 53 percent only a year ago... |
also, a little more info/background provided in this "story" on the press release
I agree totally with that. Unless I'm doing research, or hunting for a specific topic, I never use a search engine.
I have my bookmarks, and I go to those sites.
Yep, I have many many sites that are my 'favorites' and I use those all the time.
But - when I want something 'new' I always use a search engine. ;)
A potential visitor still have to find a site the first time, before he can bookmark it or somehow make note of the url -- and search engines are still the way they're doing that. Then they're getting "sophisticated" (from the itworld article) enough to bookmark the site instead of doing the same search again the next time, I guess.
On the other hand, there are still plenty of people typing their search queries into the address bar and plenty more typing "http://www.example.com" as a search engine query -- they're not all becoming sophisticates. :)
The above just confirms why we all have seen continuing declines in the total number of search engine referrals the last 4 years.
Now this all makes sense. We have always had 65% plus referrals coming from our own site.
yep.. retaining visitors and repeat visits are key. The one-off marketing web site only survived because of opportunistic search engine referals. Their time has gone. Its the end of the era of spin.
Given the recent Business Week article that now some sites are actually (gasp!) making money, is more evidence that web publishing has moved from adolesence towards maturity. Salaries are far more sensible, the mystique and novelty has worn off, dreams have been exposed for what they really were, and now its down to the hard work.
We need SEs to get the first awareness, but if a visitor leaves without leaving some sort of message or bookmarking, we consider our site a failure. Its now more important what is on the site, than getting people there.
Only substance will survive.
>We have always had 65% plus referrals
>coming from our own site.
Net average is somewhere between two and 3 clicks per user on a site. That means 66% of the referrals SHOULD be coming from your site.
Just nit picking, but "the majority of Internet sites worldwide are reached through direct navigation" is misleading. The majority of websites worldwide are rarely reached at all. :)
The majority of websites are a dying, rotting, unlinked miasma of disinterest created and promptly abandoned.
"a majority of users now go directly where they want to go"
That's a much more accurate statement. :)
This proves how important brand building is. You want people to remeber your brand (web site) and impress them making them return customers (visitors).
IMO this mostly emphasizes the importance of linkage beyond the PR concept.
I just see this happening on a new site, where the most qualified traffic comes from a fantastic link froma related site, a site which is highly respected by the targeted user group.
It's important to note in the story that search engine usage is actually up, and significantly up, from 8 to 13 percent.
Same thing happened last year. StatMarket put out a release saying direct navigation was up and showed that by comparing it agains the combined traffic of link clicks and search engine use. However, when search engine use was broken out, it turned out there hadn't been any drop at all. It was still 7-8% in 2002, as it had been in 2001.
So, the real story surprising me that for the first time in three years, search referrals have gone up -- and in fact nearly doubled.
Of course, the main takeaway remains the same. Folks find your site initially through search engines, and if they like it, they keep coming back to it directly -- assuming you made a good first impression.
These sorts of metrics are never as straight forward as they seem. I wonder what their counting as a visitor? Search engine spiders and other robots while certainly have an effect on this statistic if their requests have been included.
If we filter out this type of activity and just focus on what we believe to be human activity, we see about 75% to 80% of visitors originating from the search engines for the sites we're responsible for. But then again, these are optimised sites.
bullsh... it is :)
as someone already mentioned by counting people that use favorites as "direct hit" they destroy any possible calculations.
- those sites get favorites either by SE
- or by following sites from SE
so i think there is still 80-90% of all hits "originating" from SE
There was a good discussion here
Basically there are lies, damn lies and log analysis.
My own view is that at the end all you can compare are sets of data that have been both gathered and analysed in the same way. Even then, all you can compare are trends, not absolutes.
Brett >> The above just confirms why we all have seen continuing declines in the total number of search engine referrals the last 4 years.
One of my own unscientific reason for a decline in referrals is an increase in competition. Google and Inktomi now index several billion pages, up from several hundred million pages only a short time ago. More pages indexed means more competition in the SERPS. Every time I see a jump in the number of pages indexed, my referrals from search engines drop.
Newbies coming on the web should keep the search engines humming, but it doesn't take a rocket scientist to realize that as the web matures more and more people will be going directly to their already-identified favorite sites.
I think you also have to consider that many sites that people visit are ones that they visit every day. I remember the website names and so I can type them in. The only reason I use a search engine is if I'm looking for something unrelated.
But I tend - and I've spoken to many others within the industry - who use Google as bookmarks for their 2nd tier of websites.
First tier - 10 sites (including WW!), that I visit everyday without fail.
Second tier - 100/200 sites (maybe more!), that I know but find it quicker to type 'company name' into my toolbar and click on the result - usually right - to navigate to the site. Especially with the toolbar, I find this quicker than digging down through favourites.
Third tier - sites that I have visited, found useful, and categorised by type into my favourites - using my favs as a directory on a particualr subject is the way I use them the most.
Am I alone in my use?
|Net average is somewhere between two and 3 clicks per user on a site |
So with an average of 9 to 10 clicks anybody should be happy enough, shouldn't he?
>The above just confirms why we all have seen continuing declines in the total number of search engine referrals the last 4 years.
My SE referrals have increased dramatically in the last 4 years (over 1,000%). It depends upon what your site offers. Any site offering regional information or regional products is unlikely to be known by the majority of your target market (outside of that region). For these sites SE's are critical.
Example: A site I launched in Novemeber is now getting over 350,000 visitors per month, purely from "free" SERPs. It offers products sold on a regional basis, people from other areas have no way of knowing the domain names of suitable sites. No sites advertise on a national basis as it is not cost effective to do so.
The product this site sells is purchased on average every 7 to 10 years, even if the prospect bookmarks this site they will probably replace their PC by the time they come to look for the product again, and lose the bookmark.
If overall SE use has increased in the last 4 years, someone is getting those visitors! If it is not you, then I have to conclude you are losing out.
As for the original topic. Are most sites reached by direct domain name? It depends upon how you count. I do 300+ Google searches per day, I go to Google.com more than any other site and I do it directly. So if you count in this manner the answer is yes. But the sites I access from Google are typically 75% new to me.
Overall I access the same sites repeatidly more often than any others, but the web is a marketing machine and what about the 75% of new sites I see each day?
According to WordTracker Home Depot is searched for 35,000 times a day.....kinda odd since its domain name is highly advertised and totally obvious! "Google" is searched for 143,000 times per day, now that does require a whole new level of understanding ;)
I don't think any of these stats have much bearing on address bar use. Nor do I see Net surfing as a dying sport. On the contrary, as they say...
Search engine [read Google] use, as Danny and others point out, is up. Impressively so. Traffic from Google has doubled in the past year as MSN and Yahoo! have slipped to obscurity and AV into a spam-crazed oblivion. My business sites still average around two-thirds search engine traffic and the rest local referrals, indicating a swift path to the order form. It looks good.
Though it seems contradictory, search engine referrals being up in number does not invalidate findings suggesting a far greater percentage of people going directly to sites of their choice. Why? Is it because we've matured to such a degree or are there other factors influencing this pattern? cornwall points out that in "...the end all you can compare are sets of data that have been both gathered and analyzed in the same way. Even then, all you can compare are trends, not absolutes."
Right. One of the trends to have really slammed the Web over the past two years has been personal publishing, be it through Livejournal, Blogger, /., Movable Type, or whatnot. The influence of this phenomenon is greatly underestimated by most looking at the 'figures'.
Say I've a weblog with a link list of 50 weblogs [blogs] and sites I . I seldom need go anywhere else or refer to my Favorites. The great thing about blogs is that they're link saturated. There's no need to search for a damned thing. Everything's already linked. I've no need to remember the URLs for newspapers, zines, etc., because if the story's good, I need only follow the link from Joe Soap's blog or one of the myriad aggregators out there.
And, remember, Iím one of around two million people using a Web page of my making to hit on up to 100 sites. The potential for traffic thatís not engine related is mind boggling.
Yet, this is where the engines come in and play their part. I still use Google as my home page. Why? I generally want to read around the story or, in generating a story, want information and backup data. I now use Google to search on a far wider array of terms than ever before leading perhaps, to an increase in Google referrals to visible sites.
The distinction here has to be between personal / academic and business sites. With some two million Web users publishing link-heavy content to their own sites, their influence is going to be massive. They will travel inside their communities and referral hubs by page-generated links for a long time before they duck out to do an engine search.
Bam. Search engine referrals as a percentage of Web traffic goes down. Bad for business? Nah, I don't think so. It's bloody good for business. People buying something still use the engines. They look around. Engines will always dominate straightforward retail sites in niche markets. These sites refer on one-hit wonders and passersby.
Increasing numbers of searches and search falling as a percentage in the referrals stakes is a good thing. It shows the Web's maturing and more people are functioning 'in' the Web and traveling 'through' it rather than putting 'up' sites 'on' it. In other words, the Web's become more of a place wherein people interact than a cheapjack catalog on which we find the junk of yesteryear. Contrary to the title of this thread, web surfing has increased greatly and is offering a quality ride to all who dip their toes in the water.
In the long run, that's going to benefit us all. Hugely.
Last point. The searches coming up on small, single-page sites [where 100 page views means a good day] exercise the engines. Google has come a long way in refining searches throwing up blogs in the results [and those results are meaningful], but they've a way to go yet. A wider debate centers on whether personal publishing will, in the long run, improve content on the Web or drag it down [and, with it, the value of search]. But that's another story and another [small-world] theory...
I am a unsure how to interpret the logs, specifically the Top Referring Sites in the log.
What does "No Referrer" mean? Is it where someone types in the domain name, or is it where someone has the site bookmarked in their browser?
And how about the name of the site itself, as the referrer? "mydomain.com" Is that where someone types the domain name directly in?
|Especially with the toolbar, I find this quicker than digging down through favourites. |
I'm glad you mentioned this, JamJar. I've just recently registered the fact that I rarely bookmark anything these days - the search engines do a good job, and they tend to be more current than my bookmarks. So that's where I turn first for anything outside my "top tier".
In fact, I recenty rebuilt my home system and cleaned out thousands of bookmarks, leaving me with a very stripped down and lean bookmark file. After two weeks or so, I haven't missed a single entry.
The good thing is that, although people will browse less, they will buy more.
See half-way down this fascinating report from the UK government here [beyondbricks.com].
|What does "No Referrer" mean? Is it where someone types in the domain name, or is it where someone has the site bookmarked in their browser? |
I think you will find the answer to this question in the thread referred (no pun intended) to by Cornwall on page 1 of this thread:
Out of interest the cookie-based client side tracking that we use calls "no referrer" traffic "direct access" which might be a better term, in my opininion.
One of the inssues addressed in the aforementioned thread is the matter of how much traffic is cached by browsers, proxy servers, ISPs etc. We find that across the sites we are tracking that the cached figure is approximately 20%, although on some easily cached pages it rises to more like 40-50%. So, theoretically at least, log files could be up to 100% wrong before you start!