Compiling keyword lists is an integral part of search engine marketing. It’s arguably an integral part of niche research too. But if you’re relying solely on ‘keyword tools’ to compile your lists, you may be missing out on some significant opportunities.

By their very nature, tools that track the popularity of keywords are retrospective, rather than forward-looking. On that basis, if you’re able to identify terms about to become popular - or jump on terms that have just become popular - you may be able to achieve some quick, significant wins in terms of generating and converting traffic.

Therefore a sound approach to keyword compilation is to use a good tool to identify the quality words in your niche, as well as use a ‘nose for news’ to identify promising words.

MarketingSherpa recommends using a two-step approach where you begin with proprietary research to compile an initial list, and then use a reliable tool to generate a more finely tuned list. To compile the general list:

1. Consider your own product names, terms used on your site, and internal site search query terms.
2. Look at industry marketing materials, press releases and similar content.
3. Research online forums, blogs and social media sites for what words your target audience is using.
4. Perform competitive analysis of how other companies are positioning their sites and search ads.

After your initial research is done, the next step is to use a keyword research tool to fine-tune your overall list:

5. Use the tool to generate additional, related terms.
6. Check your existing list for estimates of search volume and competition ie. for pay-per-click (PPC) and organic search engine marketing purposes.
7. Look through historical search queries for any unusual search syntax to help you generate new search phrases

Source: MarketingSherpa, “Special Report: Online Keyword Research Guide: 5 Tips & 9 Useful Tools”, MarketingSherpa, Jun 26, 2008

David Szetela, writing in Search Engine Watch, offers some excellent tips for interpreting pay-per-click (PPC) advertising campaign data.

He explains that the most crucial data points are the conversion rate and cost-per-conversion, and that if the underlying aim of the campaign is to generate profitable sales, then these are much more important than such metrics as the click-through rate (CTR) and cost-per-click (CPC).

Yet, he makes a useful distinction. The CTR is still a useful indicator of how keywords and ads combine to generate a click, while the conversion rate measures how well keywords, ads, and the landing page combine to get a visitor to take the desired action (e.g. clicking on a link, opting in to a list, buying a product). The cost-per-conversion, meanwhile, measures the price paid to get someone to both click on the ad and take the desired action.

David explains that the results generated by a campaign can not only indicate the performance of the campaign, but can also highlight how well or poorly the campaign has been structured. For example, if the CTR and conversion rates vary widely within an ad group, the keywords in the ad group are not as related to one another, to the ad copy, or to the landing page (in the case of significant conversion rate variance), as is ideal. That indicates the keyword group should be split up into tighter groups.

Another tip is that when a particular keyword has a relatively high search volume and a lot of conversions at the right cost-per-conversion but a relatively low CTR… then it’s probably a good idea to carve off that keyword into a separate ad group - maybe even its own ad group - so that a more focused ad can be written.

On the other hand, if a keyword has a high CTR but low conversion rate, this may signify the need to better align the landing page with the PPC ad copy.

David Szetela makes an interesting point about keywords with low CTRs (e.g. less than 0.5%) and low (but not zero) conversion rates. While it may be tempting to scrap such keywords, he suggests that any keyword that produces a conversion is worth keeping. In this case it may be a matter of tweaking the ‘match’ settings and/or add some qualifiers to the keyword list (e.g. negative keywords) to improve its performance.

Check out the article for more tips for analyzing your PPC data.

Source: David Szetela, “Reading the Tea Leaves: Interpreting Keyword Reports”, Search Engine Watch, October 13, 2008