Search engine relevance and keyword selection.

Posted on 10th November 2007 at 10:22 am by Manley

Relevance, or at least user perception of relevance, is of massive importance to search engines, but there is still a limit to what they can glean from a user's search terms.

How will you optimise your pages to get maximum conversions from your traffic?

We managed to pour hot tea all over our baby last night, (there will be more than a few hospital visits but she will be fine) which serves to highlight two important issues:

1. Babies can grab your cup much faster than you can stop them, so hot drinks and babies do not mix and, no matter how much they cry, they have to stay off your lap.
2. High performance keywords are not always as obvious as one might think.

If you have ever been involved with a search campaign, be it in-house or through an agency, then you will have needed a keyword list of some kind.

You probably did not build the list yourself; it is, after all, tedious and difficult work. The users you are targeting will generally know very little about your market, many of them will be unable to spell and some of them will phrase their search in the form of a verbal question, à la 1998 Ask Jeeves. There will be long-tail opportunities that are missed, too many vanity keywords and by four o'clock all you want is a couple of Garibaldis and a cup of tea. It is not due until tomorrow anyway, is it?

I would argue that keyword selection is the most important step in any search campaign. Even before selecting a domain, the researching of keywords should have produced a comprehensive and accurate list of the search terms being used by potential customers and have indicated which of those keywords or phrases are most likely to convert.

All too often I see sites spending exorbitant amounts of time and effort chasing trophy keywords with almost zero ROI. Turning up in searches for some prestige keywords is not necessarily a bad thing - brand exposure is an important part of on-line marketing, but if a car manufacturer is looking to rank for [car] (and you will note that there are no manufacturers being returned on the first page of Google) then they are never going to see enough conversions to justify the expense.

The best example that I have been given of this is [life insurance]/[life assurance] and, rather than try to better it, I am going to plagiarise the analogy.

Insurance pays out on an event which might occur, whereas assurance pays out on something which will occur. Life plans are generally assurance policies, since we are all going to die (even those policies with a fixed termination date are general offering term assurance). Accidental death or terminal illness cover is insurance, but the majority of policies taken to cover death are assurance policies.

A quick glance at Google Trends will show how few people are aware of the difference and highlights why everyone selling life assurance on-line is also targeting [life insurance]. This is a very obvious example, but in your industry there will be more subtle ones and until you have done some detailed keyword research you are missing a huge, and more often than not cheap, chunk of your potential market.

Now, how did the NHS miss that panicking parents might not use 'scald' and who would have imagined that someone would pen a track called hot water burn baby anyway?

File under: keywords

Permalink

Comments

No-one has commented so far, or all comments are awaiting moderation.

Post Your Comment

Subscribe

RSS 2.0 Feed