Posted November 24, 2007 5:18 pm by with 11 comments

Tweet about this on TwitterShare on LinkedInShare on Google+Share on FacebookBuffer this page

(GSINC‘s Gareth Davies couldn’t resist the lure of SMX London and took excellent notes while he was there. Part 1, Part 2, Part 3, Part 4)
understanding searcher needs at SMX panel profile photos

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
Understanding the needs of a searcher
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
On the morning of day two I made my way to the first session of the day. The first person up to speak to us was Gord Hotchkiss from Enquiro. Gord had a charismatic style and even though it was first thing in the morning, he knew how to engage his audience.

Gord went straight into psychological theory to help explain the psyche of a searcher. He told us that we have a limited mental scratch pad in our brain and it is important to remember this when considering the mind and mental processes of a searcher. For example when someone is thinking about buying something they have a few slots they want to fill in mentally.

Gord gave an example of buying a laptop.

Step One: User thinks “I want to buy a laptop”
Step Two: User thinks “mmm…Ok what are the best brands?”
Step Three: “What model is best for these brands?”
And so on.

At each stage there are slots to fill, and Gord pointed out that some of these slots are already pre-assigned and filled.

When asked about laptops, many people sub-consciously thought about brands like Hp, Dell, Sony, Mac etc – these tended to be strong brands that had been successful in pre-assigning themselves into our slots. We also have pre existing ideas or stories based on brands – this is called the brand prime.

We all have limited slots in our heads

We need to accept a searcher (or potential buyer) has a finite number of slots in their head and psychologists have identified this number as anywhere between 4 -7.

See illustration below for ‘laptop brands’

7 head slots illustration

Thus most consumer decisions are about getting the potential options down from hundreds of thousands of potential options to a finite number of 4 -7. The consumer is compelled to do this so that they may get closer to the purchase they are seeking.

So we have set thresholds to reduce the options. Whilst you could say it is both conscious and unconscious behaviour it is referred to as ’banded rationality’.

Gord then gave the audience an example of when he needed to buy some GPS software for a trip around Europe and for this example he told us his thresholds were:

i/ Top 3 brands – must be one of the top 3 brands he could trust
ii/ Under $100.00 – he had decided an economical price was important
iii/ Must work on windows – had to operate on his existing PDA system.

Unfortunately for Gord the only problem with his personal search criteria and ‘Satisficing’ must have been his price banding, as he showed us that the software he ended up buying sent him on a wild goose chase around Portugal!

This process of limiting the mass of options down to a number that was appropriate for both his conscious needs and his available finite memory slots was referred to as ‘Satisficing’.

Next Gord looked at user searcher behaviour when scanning search results pages – he identified the golden triangle and the way a searcher tends to scan down to find useful results then scan across left to right, to get more information about a listing that looks relevant.

the golden triangle - graphic showing top results on google and how users eyes scan a page

Gord recapped on the process of user ‘Satisficing’ in search, and gave us the example of search terms he used to narrow down options to find his GPS software.

He started with a ‘blank slate’ state of mind, so the first search he made was general and was “gps software” This is typically a research query to help build a list. When there is not a blank slate, a user tends to have a pre existing idea about brands – so the search could include brand names. For example a person might search for “hp laptop”

Gord concluded that 92% of searchers go to search for ‘Satisficing’.

This was a great presentation. Gord integrated psychology with search in an entertaining manner and many people were hooked – including moi.

Next Jonty Kelt of Double Click stepped up and outlined a comparison between paid search and natural search.

He said that ‘Paid search is to advertising as natural search is to PR’.

He stated that paid search had a controlled message and timing and it allowed a paid for placement. Natural search lacked control over the message and timing, and whilst it was not paid for there was still generally more credibility. Jonty told us that the top of the natural search results was still seen as the benchmark for users.

Jonty then addressed the purchase funnel for search via an inverted pyramid principle (see below). He showed how an initial number of 10,000 users online could be filtered down to a 1% conversion.

Users went through a series of stages in the purchase funnel, and at each point the number was gradually reduced.

The 6 key stages of the purchase funnnel

1. Initial brand awareness
2. Brand engagement
3. Product discovery
4. Consideration and options
5. Product selection
6. Sales

A sale can result from a user that has travelled across multiple channels and so the route could be dictated by a range of advertising. This could include banners, affiliate ads, pay per click advertising, organic search listings and so on.

the purchase funnel illustration

Slow sales cycles

Jonty then examined the travel market and showed us that only 10% of purchases come from the first session that the user arrives on a travel website. 90% of sales happen after the initial session and 45% of purchases happen over a 4 week period after the initial session!

Travel market eight week buying cycle table of data

[Source data: comScore Media Metrix data run Dec 05 – Feb06]

After Jonty’s interesting presentation Sara Andresson of EMEA looked at the importance of the searcher’s intent and how this related to advertising.

She told us that the main reasons people engage in search engine marketing was primarily the following:

1. Brand awareness
2. To sell products
3. Generate leads
4. Drive traffic to raise revenue from online advertising
5. Generate leads others will close
6. To provide informational or educational content

Sara illustrated the fact that searchers show you what they want from the search phrases they use and she took the example “auto insurance”. She talked us through how it is possible to segment the search phrases and group them into themed searches which will give a good idea of what the market is looking for online.

So for the auto insurance market the searchers were grouped as follows:

48% of people were searching for quotes
Example = ‘auto insurance quote’

27% of people were searching based on price
Example = ‘cheap auto insurance’

11% were looking for vendors or companies
Example = ‘auto insurance companies’

6% were looking for functionality
Example = ‘insurance auto auction’

4% were looking for location
Example = ‘California auto insurance’

4% were general searches
For example typing in ‘auto insurance’ or “insurance+auto”

Sara then explained how searchers have different mindsets and could be broken down into 4 groups

Mindset of web users graphic -a table of data

Mapping Keyword to the buying cycle:

Similarly to the model Jonty presented, Sara showed us how the stages of intent for a user could be seen in the keywords they used. There were 4 stages and example keywords could be seen for each. For this example Sara used the keyphrase ‘crm software’ (see table below).

keyphrases search process - graphic of a table of data

Last up for this session was Tom Ally from

Tom addressed a range of data, based on a study in early 2007. The study was of 2,000 users age 18-64. 50% of those surveyed were male and 50% were female.

First Tom asked: Who are they?

– The average searcher was 33 years old
– Most searchers are employed with some college education
– 40% have graduated from college.
– 60% lived in a partnership
– 69% have incorporated the web into their daily life

More time was consumed being online each week than any other media.
– The average was 18.8 hours a week with TV making up 14.5 hours a week.

– An interesting fact was that web access from home is greater than from work
– 90% of people had broadband access.

What are they searching for?

– UK searchers are serious online shoppers
– Shopping is the #2 activity on the web after email (which is #1)

– The general trend was that people want search engines to be “answer engines”.

In the Q&A segment for this session there were some interesting points raised so I have included them here for anyone interested:

Q: How do search habits differ between men and women?

Gord told us that traditional psychology will tend to suggest that women are fuelled more by emotion and men more by fact, and whilst search engines could be seen as predominantly ‘male tools’ male and female search intent is very similar.

The main difference comes not so much with search behaviour, but more in the way websites are used. Women tend to spend longer on web pages than men. This is could be due to women having longer attention spans than men.

Q:Why don’t search engines have more paid listings on the results pages and push natural search further down?

In response to this question the panel were in general agreement that people are primarily seeking relevancy in search results and if there are too many paid ads on a web page studies have shown that it doesn’t always equate to more revenue for the publisher.

The reason being is that more ads usually mean less relevant results and whilst people could technically have more adverts to click on, there is a problem that they tend not to return as much to the search engine.

So more paid adverts mean more will be clicked on in the first instance, but revenue tends to be the greatest when balancing adverts with organic listings. This is the reason why market leaders Google continue with their model the way it is.

Baidu in China was one exception to this rule because virtually all the first page listings on Baidu are paid for sponsor listings (with organic listings pushed down to the second page). However the Chinese market is different to the UK and US and Baidu itself tends to attract lots of users searching for ringtones and free downloads. Essentially it caters for a different type of niche audience.

Related link: Baidu –

This was a great session that departed somewhat from traditional SEO and link building etc. It was refreshing to get another angle on search and there was certainly some useful eye-opening and engaging content presented.