March 08, 2007

What is the Optimal Balance of Paid and Organic Search Traffic?

In the UK, we recently launched paid and organic search terms, to show the mix of paid and organic search traffic going to websites and to show the specific paid and organic terms. In this post, I'll share some of that data for the top 10 Insurance and Appliances and Electronics websites in February, based on share of UK Internet visits. (For those of you that attended yesterday's Mad.co.uk conference - below is the data I promised!)

The following two charts show the market share of UK Internet visits going to each site, within the selected category (i.e. Insurance or Appliances and Electronics) in February on the x axis. The y axis plots the share of upstream visits each websites received from search engines in February. The size of the bubble is based on the percentage of search traffic each site got from paid listings in the four weeks to 3rd March 2006.
SEM Quadrants - Insurance.png
SEM Quadrants - Appliances & Electronics.png

As you can see, there is a wide variety in the amount of paid search traffic to leading retailers and insurance providers. The percentage varies from as low as 5% for BUPA, 9% for Dabs and Pixmania to as high as 57% for Churchill Insurance, 56% for Confused.com and 53% for T-Mobile. Even direct competitors show very different results with 50% of search traffic to Currys coming from paid listings compared to only 18% for Comet.

Churchill and Confused figure quite highly on the paid search front, likely as a result of competitiveness of financial products. In particular:
1. The effectiveness of organic listings - as the first organic listing will be the fourth on the SERP (search engine results page) financial services firms may feel forced to compete among the paid listings.
2. Competitiveness for a finite number of keywords. For example, in the four weeks to 3rd March 2007, Hitwise captured 2,490 search terms that included the keywords "car insurance". The top term, "car insurance" accounted for 29.29% of the volume of those search queries. Contrast this to "mobile phone" with 4,280 search terms, and the top term "mobile phone" only accounted for 2.86% of the volume.

In answer to the question in the title of this post - "what is the optimal balance" - it depends. It depends most on what works for you. Test, monitor and adapt constantly and continue with the tactics that yield a positive return on investment. It also depends on the structure of your website (how search engine friendly is it?).

As I've said in a previous post on the topic of paid and organic search traffic: "The lesson for marketers is that there is no hard and fast rule for the right mix of paid and organic traffic. Instead, marketers should examine return on investment from specific campaigns whilst constantly testing and adjusting search campaigns."

Posted by Heather Hopkins at 12:31 PM
Posted to Search

Trackback Pings

TrackBack URL for this entry:
http://weblogs.hitwise.com/movabletype/mt-tb.cgi/480.

Comments

More interestingly, it begs for the question - for those that have high market share but a low search contribution, what is driving their traffic instead - online display advertising? people typing their URL in directly because of brand recall? or offline advertising?

Posted by: Robin Grant at March 8, 2007 04:09 PM

Robin, This is an interesting question. Let's take the example of Moneysupermarket.

There are two things to consider.

1. Moneysupermarket is a larger site than the others on the diagram. Just because it is getting a relatively smaller share of its own traffic from search doesn't mean it is getting less traffic overall.

2. Other traffic sources making up the difference.

In response to the first issue - we can look downstream from search engines to the top Insurance websites. MoneySupermarket received more than 2x as much traffic downstream from search engines compared with The AA in February.

People sometimes get confused between upstream and downstream so let me just make sure the distinction is clear. In the scatterplot diagram in the original post, I was reporting share of upstream UK visits to each of the Insurance websites. In the figures quoted above I am reportig downstream from search engines to these same sites. The first set of figures show the relative share of traffic going to each site from search and the second is a more absolute measure - showing the share of all search engine traffic that goes to each of the sites.

In response to the second point of other traffic sources... In February, MoneySupermarket received 5% of its traffic from social networks and 21% from other Business & Finance sites. The Business & Finance sites are a mix of affiliates, consumer sites and providers. Some of this represents competitive traffic (i.e. from USwitch and Confused.com). Money Saving Expert accounted for 2% of upstream UK visits to MoneySupermarket last month.

Also, TravelSupermarket.com features prominently in the site's clickstream - accounting for 2.23% of the site's traffic. This is from consumers moving among the MoneySupermarket stable of websites.

Hope that clarifies. If you need more detail - just let me know.

Best, Heather

Posted by: Heather Hopkins at March 9, 2007 01:48 AM

Heather,
Your chart, if I read it correctly gives the example of 17% market share of UK Internet Visits to the AA. Of which 30% originate from search.
I would like to know how do you limit this to just car insurance or insurance generally? The AA obviously receives substantial traffic interested in car breakdown cover, members renewing and people using the Routefinder service. The RAC also has similar options.
Obviously Churchill and Direct Line have no other options on their sites besides insurance.
Do you track an insurance related search term through the site until the visitor shows clear intent to view information about something insurance related?

Thanks,

Mike.

Posted by: Mike at March 11, 2007 10:15 AM

Mike,

Good question... In this chart I was reporting on traffic to www.theaa.com. The only sub-domain that we report separately is www.theaa.com/travelwatch, where much of the route planner traffic goes.

The chart is not exclusively for car insurance but for all traffic to the URL's listed (i.e. www.theaa.com and www.rac.co.uk.

We can isolate specific sections of websites to report the car insurance or home insurance sections separately. This is done through bespoke reporting, not our syndicated tool. I'd be happy to put you in touch with someone that can provide more detail if you'd like.

I can tell you that much of the traffic to www.theaa.com is coming from car insurance related queries. The top three search terms sendig visits to the site last week were "aa", "the aa" and "aa car insurance".

If I understand the second part of your query correctly - you are asking whether we track people through the site. Again, we do this through bespoke reporting. But keep in mind it is all aggregated and anonimised. However, we can show some really great stats, such as for visits that came on the term "car insurance" what sections of the site did they visit, where did they go after leaving www.theaa.com. The data allows you to get a pretty detailed understanding of what is going on on your competitors' website!

Hope that helps.

Best, Heather

Posted by: Heather Hopkins at March 13, 2007 02:38 AM

Do you also check what percentage of results come from Yahoo, google and MSN. Although google is the biggest search engine I read in an article last week that more searches in the UK are carried out by Yahoo than google, which I thought would have been the opposite.

Posted by: new vans girl at July 5, 2007 03:02 AM

New Vans Girl,

Not sure of the data source for the article, as there is a pretty good consensus in the UK that Google is king.

Our updated search stats are available here:

http://weblogs.hitwise.com/heather-hopkins/2007/05/panama_uk_launch_updated_searc.html

We do report our paid and organic data by search engine. If you are interested in learning more, let me know and I can put you in touch with a someone who can provide more details.

Thanks, Heather

Posted by: Heather Hopkins at July 5, 2007 04:07 AM

Post a comment




Remember Me?

(you may use HTML tags for style)