The Digital Lookout

Search engine land reported today the release of some new functional by ask which allows a user to upload their own personal background image for the search engine.  The functionality to add a skin to the background has been available since last year but this was only for predetermined images and wasn’t customisable.  I like the idea of customising the results page and this is a much simpler solution than Google’s which involves xml information rather than a simple image upload.  It is also much more flexible and interesting than msn and yahoo’s offerings which only allow the selection of different colour palletes for the page.

This functionality is just another stage in the battle for search engine supremacy but also for loyalty within internet users through added value.  Yahoo had this a long time ago through positioning itself as an information portal and one stop shop for your internet needs (email, news, sport, search…) a similar position taken by MSN.  Then Google smashed this with its simplicity and accuracy of results.  But even the big G has recognised the need to give people more and through iGoogle struck a balance between information on the page and usability by allowing the user to choose which information feeds they received.  The issue at the bottom of all of this is keeping people using your page/engine, setting it as their homepage, and a base for all their online activities.  If they can use your site for everything they need online whey would they go elsewhere?  The longer a user in on your site, the more searches they do, the more ads they view, the more ads they click, the more money you make! Simple.  Expect a lot more releases like in this in the next 12 months as the battle continues.

It’ll take a lot more functionality for Yahoo, Ask or MSN to catch Google but I do know people who now use the Yahoo homepage as they prefer it to Google so there is some movement going on.   You can check out the Ask function on the US site here, it is not yet available in the UK.

I had my first look at the Wikia Search alpha today and I have to say the results are absolute pants!  To be fair to them the people at Wikia do say the results won’t be great at the moment as the basis of their engine is that of user reviews and not so much algorithmic search, hence results will improve rapidly over time as listings begin to get scored by users.  I have to admit that I like the idea of a user ranked search engine, after all, how many websites do you come across which have absolutely no relevance to your search phrase? (my blog ranks rather highly for “search pornsex” for example!) But not only that, a user can make more judgements on things like usability and site layout than a search engine spider which should further help the best websites rise to the top.  The process appears relatively simple, hover over a result and a five star scale will appear allowing you to score the result, this will then be used along with the algorithmic properties to determine a websites position.  This will be wholly reliant obviously on users picking up on and participating in this ranking process so I will be watching with a lot of interest how the results improve over the coming weeks.Aside from the standard results there is also going to be a section at the top of results reserved for “mini articles” on each subject.  According to Wikia “These will vary in purpose according to the circumstance, but the primary uses will be:

• Short definitions
• Disambiguations
• Photos
• See also “

Generated by the users these will obviously take the same form of the Wikipedia pages and will undoubtedly include some Wikipedia content for sections yet to be populated by the new system.  Wikia Search undoubtedly has the potential to become the most relevant search engine but the worry, as has been the problem with tagging sites such as digg in recent times, is that people begin to play the system, creating alias accounts to boost their own contents ratings and therefore rank, totally devaluing the whole platform.  If Wikia Search really does become the next number one contender to the big G then the temptation to find a “quick win” within its system will grow stronger in line with its visitor stats.  At present I believe Wikia plans to get around the duplicate account problem by basing its user on IP address but that doesn’t sound like to much of a robust system to me and I cant imaging it will be long before the spammers have an easy way of beating it.

I may have sounded negative in this post but I honestly hope Wikia Search succeeds, I hate the dominance Google has on the search market.  I also love the thought of users producing the search results rather than a piece of software.  On this initial offering I think there is a long way to go with the next big pretender.

The personalisation of search results is well documented and evidence of its arrival is clear to see particularly in the Google results. I have blogged in the past about the differing search results based on search history, the listings of site visits and also the use of IP information as a targeting tool.

But what does this all mean to advertisers? What impact is this going to have on your search marketing activity?

Position Variances

If the search engines are going to begin giving prominence to previously visited sites then you can expect variance in positions with then search engine results. For a user who has visited your site before you could be ranked top where as to a new user you could be lower down the results. This effectively is the same model as is used in the PPC listings with the metric click through rate (CTR) however this would occur at the individual level rather than the keyword.

If this begins to occur it will create confusion with advertisers as you will there will be no accurate measure of what position you are actually ranking in. This will make the performance metric of position redundant.

We are seeing this already in the PPC market with one user seeing an ad in a completely different position to another based on their search history. This makes managing a campaign much more complicated and needs to be fully considered when devising strategies.

Advance Targeting

On the positive note, this could, if used intelligently, allow advertisers to target their natural search campaigns to their key demographic. Although the basics of SEO will be needed to achieve a decent position in the first place once this is in place effective use of creative and website copy could lead to a increased performance for a particular segment and so increased positions. This will be dependant on the extent personalisation takes effect and will need a very clever implementation but should be possible.

I’m sure there will be more effects which come about as a result of the continued personalisation of results with in the search engines but for now these are the key two. What this space, it’s going to get increasingly complicated!

On a Google search yesterday I spotted something I had never seen before which is further evidence of the increasing personalisation of search results. On a search result for the AdWords help centre Google told me how many times I had visited the page and when I last visited (see screenshot). At the time I was signed in to my Google homepage and so it would be linked to my session I am sure but it is a further indication we are on our way to a fully personalised SERP. It will be interesting to see if they begin to use this as a CTR metric as with the PPC model and use it to rank the results in the same way.

As the evolution of the search engine results page (SERP) pick up speed advertisers need to begin asking the question, what does it all mean to me?

The advent of Google’s universal search and the inevitable following suit of the other search engines will signify a new dawn in search engine marketing. I have listed below a few things which I think will be impacted and that must be considered when optimising for the future of search:

PPC: Logic suggests that with the introduction of more information onto the SERP that the number of paid links will be reduced. But then they are the main income source of the page, so will they? I think the likely hood is that the number will be reduced to 6-8 listings per page (as with the new ask.com pages), enough to free up some inventory but not significantly reduce earnings. In fact, this sort of reduction could drive up the price of first page exposure as everyone clambers for prime position so there may be no reduction in revenues at all. This could mean advertisers see a drop off in traffic as the page the are appearing on begins to drop, and a rise in costs as they fight to achieve exposure.

Dependant on the positioning of the paid listings in the page there could also be changes in click through rate. If you are adding in additional elements into the does this mean the clicks are diluted between them? I would suggest this is down how the information is arranged and the prominence each is given.

SEO: Again there could be a drop in the number of natural listings which appear in the SERP as additional inventory is introduced. This makes the importance of front page exposure even more valuable and harder to come by!

There may also be a reduction in the text which is displayed for each result in order to fit in more listings. If this is the case it would become extremely important to ensure that title tags are not only targeted to the keyword but are enticing to the user.

Additional Products: With the addition of “other” results into the SERP it will become increasingly important to take a holistic approach and incorporate all of them into an effective search campaign. Image search, maps, product feeds, book listings, will all become important sources of traffic as their prominence and exposure increases.

There is a big win to be had for the advertiser which embraces the changing face of search and optimises for all elements of the page. Rather than a threat to each individual element this should be seen as an opportunity for front page exposure on numerous fronts. The companies/advertisers which adapt with the SERP will prosper, those that don’t will be left wondering where all the traffic went!