Content, Content, Content

SEM
Search Engine Marketing is one of the ways you can profile your web site. Basically, it is like an outer envelope in a direct mail piece (forgive me, web guru's, I know that sounds reductive, but it's a pretty perfect analogy). The fun of it is that the outer envelope can have many different shapes, sizes and tag lines.
But the success of SEM is dependent on some pretty simple principles.
First of all, you need to clearly define your campaign goal. Are you looking for sheer number of clicks, length of time spent on the site, reduced bounce rates, number of page views?
Secondly, you need to have a good idea of what people are searching for. It's smart to start an SEM campaign with multiple threads, because that helps you unravel patterns. Then you're able to use real world knowledge to set your campaign. Because the web is real time, the first few weeks of the campaign should be well monitored, giving you the opportunity to change things on the fly.
Thirdly, brand recognition boosts results. Coinciding campaigns, media presence and buzz all increase web activity. SEM is a part of the mix -- not the foundation of a marketing plan.
Finally, and I think the most impacting, the content on your site determines ultimate success.
Some SEM specialists trick people to the site. So they use ad words that can be loosely attributed to web content and are highly searched. Which means the ad comes up frequently, has a high click rate -- but, unfortunately, often has a very high bounce rate. When the viewer clicks on the ad and goes to the site and sees that the content is not really what they were looking for -- its takes about 1 second before they click away.
This Christmas we launched a number of SEM tests.
Two profiled new Christmas symbolic gift product lines. Both were versions of the "goat, chicken" offers from international development agencies. Both were really new in the space -- so they were unfamiliar to the mass market. While both had simple offers, they also were more complicated than the buying 5 chickens. Neither were well supported by media, communications or additional marketing activity. Both were evaluated by the number of sales generated. While the final numbers have not been completely evaluated, neither received stellar results.
There were a couple of reasons. Very few people were searching for these specific products. They didn't know to search for them. Secondly, they were not hot products for Christmas. They did not have mass appeal. Both sites had sufficient content, but people were not that interested in buying the product.
We also launched a campaign for a custom home builder. The site was already performing well -- with an average viewer staying on the site for 6 minutes and viewing 10 - 16 pages. We thought that the rate would plummet with the SEM campaign -- simply because the site would draw more visitors.
But no.... the length of each visit remained high. The number of unique visitors increased dramatically and the number of page visits continued to be high.
The site is fat with content. The ads profiled that content. When the visitors arrived at the site, they were not disappointed. So they stayed.
We set the initial campaign fairly low -- wanting to assess the strength of the various campaigns. It looks like we will have to up the spend a bit, because the click rate is so healthy.
SEM that is not supported by a site that has great content is like sending people a direct mail with a great teaser and when they open the envelope it's empty.