MSN Asks for Help & Feedback with Search Algorithms

Source Title:
We are here and we are listening
Story Text:

Strange thread at WMW started by the self styled "msndude", a MSN employee with the grand total of 33 posts there in the last year and a half. Is this a genuine cry for help, PR spin, the last twitching of a dead search engine? What do they expect to get back?

We are here and we are listening happy! ....

With regards to spam we are constantly working on this and we have improved and will continue to. Please send along any private relevance feedback via stickymail.

Also, there are other folks that seem to say MSN Search is no good. That may be fair feedback, however, would love to get some ideas on where we need to improve most -- is it spam, relevance or another area altogether?


Seems like

Seems like MSN can use the help...

The problem isn't

The problem isn't determining "spam" but simply recognising relevancy - MSN search seems very on-page focussed, which is precisely the restraint Google broke out from and made it so successful.

Yup, MSN need a new algo

I've done a couple of tests using my standard search strings, and 2 big things stood out for me

1) their linking dataflow is rubbish. Way too much on the outbound anchors
2) Their hub / authority model is worse. I suspect this is mostly a function of their lack of long term data. The best move they could make, IMO, is to buy up an old, but cheap search engine with crawling logs that extend back further in time than their own.

Without the historic data, they just can't put a proper analysis together, which is why they are over reliant on on-page factors at the moment, I think


You could just tell them that the average MSN search user is so computer illiterate they haven't managed to change their homepage to Google :)

So it really doesn't matter how good the results are.

just put ppc in

Why bother about serving up the organic stuff.....

It's Not The Results

Relevancy isn't a scarce commodity. As I far as I can see, there's very little to seperate the major search engines in terms of their ability to deliver relevancy.

MSN's problem? There is no good reason for me to change to MSN. And even if I did, what is in it for me?

MSN need to give people a sense of ownership.

Google has ALWAYS looked at on-page factors

Where do people come up with these crazy ideas about Google? They have never stopped looking at on-page factors.

MSN's biggest problem is perception. They actually do a pretty good job of determining relevance (better than Google, even, in many cases). But they cannot shake their mother company's shady reputation. People do not trust Microsoft.

Why would the little people...

...want to help the big people, that lesson has already, I hope, been learnt! As long as the big boys are still fighting to attain a credible amount on recognition then that must be in my (little person’s) interest. Once they (big person) have cut the market up and are happy they’ll charge us to use our (ooppps I’m sorry) their space.

I have to agree with Michael

I have to agree with Michael - MSN's biggest problem is perception. I think Rusty's test a few months back showed us that the difference in relevance is actually quite small. MSN needs serious work on the number of results they report and they need some time in the game to get temporal link analysis down, but on the whole, I see them doing a great job for the newest player in the game.

Without the historic data,

Without the historic data, they just can't put a proper analysis together, which is why they are over reliant on on-page factors at the moment...


"Google has ALWAYS looked at

"Google has ALWAYS looked at on-page factors"

That's not in doubt. What is in doubt is promoting a search engine where on-page factors are still the coup de grace for ranking relevancy. That's a part of MSN's problem.

MSN is young

Every search engine has to start at the beginning. If MSN doesn't pull itself up to a more highly competitive status in two years, then I'll be ready to give up on them.

Until then, I'll cut them some slack.

Especially given how much people have abused the external factors for establishing relevance, I hope and pray MSN doesn't repeat the mistakes that Google has made.

MSN must do four things

1) Listen to their $$ market. The SEM community and advertisers as a whole.
2) Listen to the searchers and capitalize on the other engine's general lack of interest in quality improvements. Develop leading edge search with new or different functionality. Not just the same ol' search like everyone else.
3) Show the vision and leadership of combining #1 and #2 and win over their share of the market.
4) Clean up their overall rep!

If they can do that then they will be successful. They spent $500 million in branding for MSN Search and it got them what? Nothing. Why? They haven't won over the support of the people where their main interest lies.

Right now, MSN Search has no presence in the market they want to play in. They need to win over the market insiders in order to gain momentum and move forward. It's not always about having a better algo. Sometimes it's about winning friends and influencing people. The only way to unseat Google is to be a “media darling”. Until Microsoft cleans up its reputation as a whole this will never happen.

Anyone want to take a guess if they will ever spend $500 million on doing that?

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.