German search traffic: real life figures

Search engine traffic

[heini]

So lets look at our logs and break it down:
Where does traffic to german sites come from:

Me:
Google.de >55-65%
Google.com 5%
T-Online 5% (Fast)
MSN 5%
Yahoo.de 5% (Google)
Lycos.de <5% (Fast)
AOL.de <5% (Google)
Web.de <5% (Google)
Metager (Meta engine) <5%
Rest is in the <1% range.

That is disregarding PPC. Looking across several sites, all fairly well propagated, some e-com, some purely info sites.

How are your average numbers?

[heini]

>Yahoo, MSN and T-Online are better than they seem to be in the logs of SEO-aware people.

Fischerländer, I was assuming that people posting here have good rankings across the board.
I suspect the webhits clientele might not be very good with Google, which may be another explanation of the differences. Plus webhits does not differentiate between PPC driven traffic and *free* traffic.
Generally I'm fairly sure Google is underrepresented in the webhits figures, and the rest consequently overrepresented.

Netnerd, we have a dedicated UK SEs forum, where you'll find lots of info. You might start here:
[webmasterworld.com...]

[Fischerlaender]

heini wrote:

I suspect the webhits clientele might not be very good with Google, which may be another explanation of the differences. Plus webhits does not differentiate between PPC driven traffic and *free* traffic.
Generally I'm fairly sure Google is underrepresented in the webhits figures, and the rest consequently overrepresented.

That's really interesting, because several people suggested to me that webhits overrepresents Google.

There is another thing to think of: A counter service like webhits is very often just built into the homepage. This means that visitors from search engines who are sent directly to a subpage, aren't seen by the webhits statistic. Engines with a link popularity based algorithm tend to show homepages as the best results. Thus they should be overrepresented by the webhits statistic ...

[heini]

Ya, but which SE does not take linkpop into account? Fast is notorious for at least sporadically overrepresenting indx pages. And MSN, with their directory listings in top slots are sure to overrepresent homepages. Actually on none (had to check in detail, but I'm fairly sure it would be an exception) of my sites the homepage is the page with the highest Google traffic.

Where I agree is that some people here may have a strong Google orientation.
But I certainly wouldn't be surprised if among the average site with a webhits counter Google positions were rather lousy.

[h_b_k]

Hi,

as I have posted months ago in FAST-forum, I am in trouble with FAST's family filter.
I have very good SERP positions with filter-off option. The filter on-results differs from week to week. And the search traffic from t-online and lycos is following this trend.

[oLeon]

As I posted before - my numbers are telling another thing:
we got from 20 % up to almost 40 % traffic from the others than Google.

Why? Because we are able to do it! Of course G is important but I doubt that itīs that big as mentioned in other postings.
Why? Oh, could it be, that itīs easier to reach top positions on Google than on FAST/Lycos? I believe so. Everyone has a focus on G, and only if one has a bit time beside this G-goal, we may stare around to see what the other SE do.
Furthermore the changes on FAST at fixed intervals are annoying but do help to have a different index all two/four weeks. For users itīs fine, for us SEOs not at all.

I agree that FAST pushes the index page. And if this is not really optimised you lose a potential traffic generator.

Nevertheless - we may not forget, that G is the biggest one we have. Doesnīt matter if 60, 70, 80 % - what important is, that G is TOO big.
First when you lose all listings, you know what I mean!

[weesnich]

The problem is that it pays more to work on google-rankings. If google has 70% marketshare and I manage to improve my rankings to a position that is (in numbers of traffic) 50% better - it will result in total to 135% of the old traffic.

Same on fast (lets say they have 20% marketshare) only results in 110% of the old value.

[hyperion]

weesnich, marketshare and competition should egalize each other - if the market share is 1/5 of google, 1/5 of SEOs will compete for the first position, so the time invested theoretically would pay of.
But @oLeon, that only works if optimizing works without paying. If, weighted by marketshare, in 90% of the places where FAST results show up, they show up after PPC-results, why should I bother to optimize? When I optimize for google, I know that 80-90% of the results will show up on google itself, without PPC-entries shown before them, and because of that, it is worthwile to optimize for google. If you are really after the clicks from inktomi or FASTs subsidiaries, it is easier to pay for them.

[Fiete]

SEOs, be careful not to mix up cause and effect when regarding traffic and optimization!

Isn't it possible or even likely, that you get about 60 or 70 % of traffic by Google just BECAUSE you optimize your site for it?

weesnich, your calculation of increased traffic only works under the condition that optimizing for a "50 %-better" ranking causes the same effort at each search engine. I doubt that this is the case. I am pretty sure that on the smaller or niche search engines it is much easier, because there are not so many others who optimize for this SE.

Probably it is worth to spend some (but not as much as for the big ones) time into smaller SE. The effect compared to the cost may be worth it.

By the way, "market share" in terms of traffic reaching your site may be interesting from the search enginge's point of view. For the site owner, the increase of ABSOLUTE traffic figures is important, not the share!

[heini]

I have opened this discussion under the assumption of people sharing numbers from a sample of sites, which rank well across the board.
Sure, most people optimize mostly with Google in mind.
But, what does Google optimization mean? It's one half at least getting the right links, i.e. external factors. The other half is basically traditional onpage optimization. While links and the specifics of PR are special with Google, rankings in all other major engines are likely to profit from that too. Basically the same holds true for onpage stuff.
In consequence I don't see the argument about only optimizing for Google screwing the stats holding true.

Given that many totally unoptimized sites tend to do well in Google, if those are old established sites, or community sites, the whole educational, public sector, and larger private networks, I don't think SEOs have a larger share of Google traffic than average.

It might even be argued that SEOs see a larger share of non Google traffic, because they actively pursue indexing and ranking in other engines and directories. This might to a degree explain the high numbers a noted pecialist in this thread gives for engines like Fireball, or MSN. Fireball rarely spiders a site which is not actively pushed into their index.

Bottom line: I would expect a larger share of Google traffic among webpublishers, non commercial sites, and non SEOed sites than among professionally promoted sites.

[Bernie]

I agree with oLeon that there could be some overrepresentation with Google because SEOs focus on it.

The google-algo is pretty reliable hence it is "easy" to position a website in Google. (I am not talking about highly competitive keywords) Whereas esp. Fast keeps being a mystery to many of us. ;-)

On the other hand the Webhits-Statistics should then show a stronger market share of MSN, Lycos, T-online and Fireball to catch up with oLeon's 40% non-google se-referrers.

The main reason I am not that much interested in the german "second league" (MSN,T-Online,Lycos,FB) is not the market-shares it is the fact that they all stepped into PPC and do more or less conceal the PPC.

The most slapping example is T-Online. Wanna do SEO for T-Online by optimizing for FAST? you better safe time and go to overture.de BTW: we have excellent click to sale rates with overture listings on T-Online.

[heini]

Sorry, folks, but the logic here escapes me.
Assuming I have an unoptimized site, rather large, with a typical mix of reciprocal links from friends or partners, eventually one or two directory listings.

What are the chances such a site does better in Fast, MSN, Fireball than in Google?

I would almost certainly expect that site only doing reasonably well, if at all, in Google. Perhaps Fast. But very likely it will not even be in MSN or Fireball's index. At T-online you have to be in the top 5 to even show up on page one, and at #2 or two, to appear above the fold on many searches and typical window sizes.

I really can't see any reason why a professionally promoted site should have an unnaturally high share of Google traffic. Rather the opposite.
I wouldn't be surprised to see a 90% or more share of Google traffic for any hobby site, which does not get promoted at all.

[oLeon]

just to hinder a misunderstandig:

certainly itīs not worth it to optimise for FAST by forgetting Google. no doubt.
but otherwise itīs a certain thing: many SEO only think about G by optimising, and cannot reach a click from other SEs.

what I want to suggest is - donīt forget G during working for FAST, and donīt forget FAST + INK working for G.
could this be successful? yes, it could. if you did a good work for G itīs a common thing to be good presented on FAST as well. I reached this aim for some clients, but it takes a lot of time.

unfortunately I do agree with you, heini ;-)

but just hold in mind, that SEO also means to provide listings in directories, and if this is done good, the promoted sites will appear e.g. on MSN at the top, too.

I donīt define SEO as "only searchengines related".

[Bernie]

heini,

I really can't see any reason why a professionally promoted site should have an unnaturally high share of Google traffic. Rather the opposite.

I'll give you one: with some of my own stuff (non client projects) I do only care for google. I can show you some logfiles with 95% Google referrers. (including google using ses)

Why? Because it can be better to just care for Google and go for the next project instead of loosing to much time caring about the rest of the market.

If you don't have clients who expect rankings in all searchsystems you have good reasons to drop msn, fireball and fast

msn: does ppc
t-online: does ppc to an extreme - where does the real serp begin? it's page 4 for some keywords now!
fireball: can be a pain in the a** with deep-indexing

plus - as I told you in London - I do not 100% understand fast - and so do admit others. ;-)

[Fischerlaender]

Optimizing for Google = Avoid technical problems like redirects or dumb JavaScripts and set some links from sites with a good PR.

Optimizing for the rest = Avoid technical problems, adjust onpage features (e.g. keyword density), set links out of the blue because FAST and the likes don't show their "PR".

My experience shows me two things:
1. It's a lot easier to get top positions for Google than for other engines.
2. Clients are only interested in Google positions.

What does this mean? I just optimize for Google and if I get some good positions elsewhere, then this a nice side effect.

FYI, I just pulled out some stats from other sites:

Google......91.5% 
Yahoo........4.5% 
Web.de.......2.3%

Google......78.5% 
Firebal......6.5% 
Yahoo........5.7% 
Lycos........3.7% 
T-Online.....3.5% 
MSN..........1.5% 
Web.de.......0.1%

[oLeon]

Clients are only interested in Google positions.

oh yeah, I forgot this detail. there are a lot of clients that are not interested in FAST & co.
sometimes I think, the clients are as paranoid as we are :-)

[hyperion]

Hhm, just checked the position in Fireball for the site whose statistics I posted earlier, actually the position in Fireball is more or less equal to Google for the two most relevant keywords (in Google, it is second and 11th position, in fireball it is third and fifth). Still, Fireball gives me less than 0.5% of the referrals I get from google.de.
So, I think this is really market specific. If you target the technically savy, you can forget about all the mainstream alternatives in germany, I'd say.

[oLeon]

I have to admit that I donīt know whether my referrer from the others than Google are that high due to the wunderful rankings on them or due to my lack of ability to be in the TOP 10 on Google by searching for generic words.

Thatīs an intersting point I have to think about...

[albert]

Sorry if my post is a little late in this thread - but I found it by accident today.

I have some figures about SE traffic for 7 sites (com domains, language english only, hosted in Germany), and they show a clear dominance of Google:

Google 62 - 78 %
Yahoo 9,5 - 21 %
MSN 6,6 - 14,5 %
AOL 1,2 - 5,9 %
all others mostly below 1 %

No PPC. All sites are well propagated and doing well. Figures cover the last 3 months.

Interesting point for me:
If searches for keywords related to those sites are in english (because sites are english-only), why are those figures more similar to the figures in this thread concerning german search engine traffic, and less similar to figures for example in US?

[albert]

Apparently not so much traffic here ...

Anybody awake in europe? Sorry, but I was curious about your experiences and figures.

[Rumbas]

Thanks for the numbers albert.

Sorry, I'm not sure what you mean here. I guess the search pattern is dependant on what type of keywords you're targetting.

Could you spell it our for us slow types? ;)

[oLeon]

I am tired...and really donīt know what you mean, albert.

If I understand it correctly, I īd assume it depends on the hosting issue.
What ccTLD is your domain?

[albert]

Sorry if my post wasn't clear enough.

I expected my SE traffic figures to be similar to international SE traffic figures.

In fact they seem to differ compared to e.g. Searchenginewatch.

But they are similar to figures I found in this thread (e.g. msg #1 and 2) concerning sites in GERMAN, I assume.

My sites are in ENGLISH (but hosted in Germany).

So why my SE traffic's pattern is similar to german SE traffic pattern, and not to international?

Targetted keywords are names of internationally known designers and their well-known objects ...

Sorry again if my question is still not clear.

[oLeon]

again:
which ccTLD is it?
.com - .net - .co.uk - or whatever...

to be similar to international SE traffic figures

What is the traffic you expected?

[albert]

Hi Oleon,

as I said in my first post:
all are com domains.

Expected traffic:
Searchenginewatch has
Google 29,5%
Yahoo 28,9%
MSN 27,6%
AOL 18,4%
a.s.o.

Their figures are based on <quote>:
>>A panel of more than 60,000 home and work surfers was measured to estimate these figures. Because a web surfer may visit more than one service, the combined totals exceed 100 percent.<<

I've found other posts in different threads at this forum with similar figures to that of Searchenginewatch.

<edit>typo</edit>

[oLeon]

as I said in my first post:
all are com domains.

Oops, sorry. Didnīt see this (just sleeping...)

1st:
Do you have listings in directories?
The figures from SEW are based on this, too.

2nd:
I assume you are not listed in UK by choosing the option "pages from the UK". If you wanted you have to be hosted in UK.
But you are listed by choosing the same option in Germany...even if you donīt want to.

3rd:
your KWs are obviously very generic - too generic. And they are relevant in all Gīs all over the world. E.g. mobilephones: the names are the same all over the world.

[albert]

Thanks, oLeon.

@1: I have 2 of my 7 sites in Yahoo. None in DMOZ.

@2: If you speak about Google - I guess I'm listed in UK as well (even if I don't understand why this is important to my question).

@3: Maybe my keywords are generic (BTW: is a designer's name really generic?). - But I have no problems to be found (very good / excellent listings for my targetted keywords in important SE's like Google, Yahoo, MSN, AOL ...). I have very nice traffic especially from SE's, and my sites do well. So I'm contented so far.

Only wondering about the SE traffic pattern ...

I thought that either the SEW figures are biased (US only), or my logs are biased (I use Webtrends).

But maybe I just made some error in reasoning ... ;-)

[heini]

Hi Albert, if I understand correctly, your figures are derived mostly from US/worldwide traffic. That's a whole different story to Germany.
We can of course compare the numbers, but I'm afraid we would need a more broad sample of US/english language traffic than just one site.

As to the figures given by Searchenginewatch, those are US only, I believe? whereas your figures are worldwide.
First thing to get a grip on would be the spread of countries from where your visitors originate.

[albert]

Hi Heini:

As you said my figures are from worldwide traffic, and SEW is US only. I know that this thread is about traffic in Germany (concerning websites in german language).

So I guess my post is in the wrong thread. Those similarities between figures mentioned here and mine may be arbitrary.

Thanks, anyway.