Search Engine Strategies -Munich (part one)

Search Engine Strategies Conference & Expo 2003 – München

The search engine strategies conference held at the Arabella Sheraton Hotel in Munich last month was my first, and I was honoured to be a speaker at the event. The attendance was quite good considering that Germany has not really warmed to search engine marketing in anywhere near the levels of most English speaking nations. There was still however a good attendance including representatives from some of Germanys leading corporations.

I unfortunately missed Danny Sullivans opening session as I had to travel from Hannover which is a 5 hour train Germany and only arrived at 10.30 am. On arrival I met the organizer (Mathew Finlay) and then it was off for the first session (the one before my own and a lunch break).

Below a short report from the sessions I attended.

The first session I attended was the one carried out by Shari Thurow from Grantasticdesigns.com on search engine friendly design. It was a very professional presentation with many examples on search engine friendly design. I confess to not agreeing with all that was said from Shari. One example being the claim from Shari that keywords in domain names do not have any influence in ranking. They do, but not in the way most people think they do, in that Google automatically gives a ranking boost for websites with domain names in them. Keywords in a domain DO effect a ranking due to the direct url links that websites may give. Link text is very powerful and if you are getting linked with just your url, and that url happens to have a keyword in it, you automatically gain a boost due to the link text. There was one other thing mentioned which raised my eyebrows and that was that Shari claimed that not only the anchor text is important but also the immediately surrounding text also provides a boost in the ranking. I have no evidence to say that is not the case but have my doubts that the there is any special boost in having surrounding text of links that includes keywords apart from the normal weighting given to body text including keywords. It is however certainly something worth trying out.
Shari did an excellent job explaining keyword prominence. How to get the main keyword rich content at the top of the page. Examples were using hyperlinks that jump to different sections in the page on a FAQ rather than just listing them one after the other. Sitemaps, glossaries and footer links are things I write about myself and I agreed with her fully on those issues.

After the session it was lunch, which myself and Stefan Fischerländer left early to prepare for our session on keyword research. There were however technical problems as we could not get online. A technician was called and the visitors filed in with our moderator Tim Cole. Our starting time came and went and still we couldn’t get online. Stefan was to start and as he didn’t need to go online, laptops were exchanged and the session started 10 minutes late with the technician in a corner trying to sort out the internet access which was important for my part (resolved by simply rebooting in the end). Stefan concentrated on the importance of keyword research and how it can make a huge difference to your bottom line and then I followed after 25 minutes. We changed the laptops over and then I started. German is not my native language and it was my first public speaking session on the theme and my first ever German language session. It was tough going and I know that due to a few nerves, I made some real howlers in my grammar. As I became more confident however, I started to enjoy myself and went into the 5 different brainstorming method for keyword research..

1. Online tools
2. Competitors pages
3. log files
4. friends and colleagues
5. internal site search function and logging.


I showed online a few of the available keyword research tools. Most of the audience I was sure already new about the overture tool for example so I didn’t spend to long on the free tools available. I then went on to post optimizing keyword occurrence checking. This is where I was hoping for some interaction and called upon members of the audience to provide their url for checking on my Topword tool http://www.abakus-internet-marketin...ls/topword.html The tool analyses a webpage and outputs the occurrences of keywords and keyword phrases split into singular, two word and three word phrases. Four of those in the audience submitted their url for scrutiny and on analysis some probably wished they hadn’t, as some of the top 2 word phrases were phrases like “about us”, “company name”, “click here” etc. It certainly opened the eyes of many present as to why they were not ranking. Their keywords were simply not on the page. All in all the session went well and I’m sure the visitors got something out of it. Although I wish next time I had more than one night to prepare and rehearse as its hard enough to do well in your own language, let alone in a foreign one.

After my session I then spent a lot of time talking to those who had stands and generally networking.

On the next day there were to be some sessions that I really didn’t want to miss. The one everyone wanted to go to was the “Meet the Crawlers” session with a real life Google software engineer no less, who was to give a presentation on the Google crawler. Accompanied by inktomi and Lycos. There was also to be a session on Dynamic websites which was something that I took a high personal interest in.

The Dynamic website session was first and it was very well presented. Although I didn’t personally learn anything new from the session, I’ sure many did. The session covered the different kinds of dynamic sites there are (template and file based) and possible solutions to the multiple parameter problems. This meant that workarounds were discussed such as archiving pages in pure html format, mod_rewrite, sitemaps to help indexation and ISAPI filters for windows servers. Unfortunately there were no examples of mod_rewrite given or asapi filters. There was a lot of however on how to make the most out of a dynamic website with as mentioned archiving pages and sitemaps.

The next session I attended was to be the “Meet the Crawlers” session which started off with the Inktomi representative talking about how wan independent study revealed that users found the Inktomi search results more relevant than any other search engine, including Google. The Google software engineers face was a picture when the view foil went up showing Inktomi in top place. The inktomi man then went on to describe the crawling routine being approximately 10 days for the full index and there was talk of course on the pay per inclusion side of things. Nothing however particularly meaty. Then Lycos had their say. This was pretty much more of the same. Although it was mentioned that the date of the file is a factor in the Fireball algorithm. Most SEOs know this already however. Then the man everyone wanted to hear came to the platform.

The Google software engineer (Michael Schmitt) was a young guy who certainly looked the part and the first thing he did was refute Inktomis suggestion that Google produced results that were not as relevant as Inktomis. Unfortunately the moderator, a professor whose name escapes me, was determined to interrupt and have his say from the very start. Quite frustrating as he didn’t say anything worthwhile and was just eating into the time Mr. Schmitt had. Introductions to Google were made with a brief history and then he got down to the way Google crawls. There is a 3 tier indexation on google. Main, Fresh and News. News gets crawled virtually non-stop. Fresh every other day and Main once a week or more. Although asked, he was not prepared to say how a website gets from a main crawl to a fresh call. Logically however you would expect fresh sites to be ‘expert’ sites. High pagerank and regularly changed with plenty of content. Understandably Mr. Schmitt was not going to be persuaded to reveal anything meaty about the search engine algorithm. He did however make some quite important confirmation on issues that have always been suspected. The Google algorithm most definitely take the following into consideration.

Link text
Size of text
Distance of keywords between each other
Keyword density

continued in part two below.

Interesting was the term size of text. Does that mean a large font size is weighted as a H1 tag would be? Or is it that he didn’t’ want to specifically mention the heading tags.
Nothing particularly new there, but its good to have at least some confirmation that I’ve been optimizing correctly up until now. He went on to say there are around 100 other factors that go into the algorithm.

Another important confirmation involved the indexation of dynamic webpages. Mr. Schmitt confirmed that Google will index dynamic pages with up to 2 parameters. But has problems with more than that.

The subject of Spam inevitably arose. Mr. Schmitt was in fact very humble and was quick to admit that German searches produced a great deal more spam results than that of English language/location searches. Apparently Google has only concentrated on effective spam filters for the English language interaces and searches, and little has been done for Google.de. He agreed that there was a problem and recognised that due to Googles failure to nip Spam in the bud early on, there are a great deal of hobby SEOs spamming in one form or another as the SERPs reveal spamming works and there is no downside. One reason given for the state of the German Google results given was that on Webmasterworld there is Googleguy who is there to act as a guide and deter webmasters from spamming. Although he has a point, I submit the major reason is as he himself admitted, in that not enough attention has been given to .de results and that the main research, testing and resources is going to google.com and the English speaking Googles. He was however brave to stand up there and admit there were problems. He couldn’t really do otherwise though. Apparently there are going to be changes soon with new filters etc. The problem is, many SEOs who follow the guidelines have heard that before and nothing has changed.

Mr. Schmitt when asked if he thought SEOs were a problem for Google, replied that in fact if an SEO shows a client how to make their site much more search engine friendly then that is a good thing. The only problems he claims are when the SEOs do not follow the Google guidelines and produce hundreds of poor quality doorway pages with redirects etc. That was good to hear.

After this session it was the portals/directories turn. Something that I wasn’t particularly interested in so I went and did some more mingling and had a long coffee break. The day was finished after my last session I attended which was a website clinic where visitors could have their site reviewed from a panel. I listened in case I might pick up something new, alas I didn’t, but the panel handled the reviews well. And then it was time to say goodbye to those who I’d met and at 6pm caught the train back to Hannover.

Thanks this is very helpful. i will follow up with my report on SES Chicago Dec 9 - 11, I will be there.

Webby, was the German SES one day?

No, it was a two day thing.
I look forward to your report on the San Diego event :-)

I was actually planning to go to the chicago show but didn'r feel lkke spending 1500 to listen to someone repeat what I already know. I would mind seeing Sullivan speak though.

-PK

Sullivan only gives the intro - a generic SEO/SEM speach.

I'll give my report on Chicago and let you know how it goes.

Its a 3 day event and I hope to pick up new stuff but the purpose is to meet people.

Great stuff, especially the Google part...



This became quite clear with florida-update. All non-english sites I know (mainly finnish, swedish or german) were allmost completely untouched. I would really like to see google taking a match with finnish grammar - it's a battle I'd bet google loose ;)

It's a battle everyone would lose isnt it?? ;-)

Thanks Webby! Nice narrative.



This is a bit dated considering the multiple algo changes that Google has implemented since the posts were written, but it is interesting none-the-less:

http://www.ihelpyouservices.com/for...=&threadid=7675