Friday, July 25, 2008

B2B keywords - an SEO insider look

Recently, we had a B2B (Business to Business, for example a manufacturer of plastic resins seeking plastic manufacturing and assembly customers) customer ask us to develop their keywords. In this particularly esoteric field, the exercise turned out to be quite a learning experience - but nonetheless we developed a good initial, targeted list of 100 words and phrases, using several tools and a good deal of manual research/analysis. Their next request was to ask exactly which were the highest priority words, and the explicit statistics to be back it up.

SEO/SEM being the inexact art that it is, ESPECIALLY when dealing with B2B, here was our answer:

"On the whole, the keywords that make sense with respect to CUSTOMER XXX aren’t showing up very frequently at all, in some tools that are able to collect statistics from search engines. These tools, as well, do not necessarily reflect the “real” universe of searches – for example Wordtracker collects from the meta-aggregators Dogpile and Metacrawler – search engines we’ve found to be less accurate and useful than straight Google, Yahoo or MSN (mainly because their results favor paid search results, and reflect their own business priorities vs. “natural” results).

Therefore, most of the ranking and sorting has to come about through manual scanning and interpreting of the prevalence, frequency and priority placed on particular words and phrases among the competitor’s sites, articles and other material found while searching around the “industry”. As well, certain subjective discretion is taken, based on SEO/SEM experience, to focus on word and phrases that are more likely to be used in search strings because of capitalization features, abbreviations, avoidance of conflicting terms, and the like.

Ultimately, initial keyword choice and ranking, especially in this fairly narrow B2B market, is going to be an initial “best guess” backed up by some degree of professional experience and to a smaller extent, standard tool output. Very precise, statistical comparisons of keywords across various contexts and test cases can certainly be done with very advanced tools – but that costs a lot of time and effort (can take several weeks, with use of very expensive tools), exceed the current scope of effort and budget.

Note also that the current state of the Internet, with respect to what people are searching for, can be manipulated. This means that CUSTOMER XXX, based on the material it produces and distributes around the web, can actually provoke more people to search for particular keywords. Consider an article on the benefits of “metals conversion”, created and distributed with professional Search Engine and Social Media Marketing help. This increases the online conversation about the topic, results in additional contribution and “buzz”, and in the end increases the search frequency associated with this phrase – all pointing back to CUSTOMER XXX.

Therefore, for this degree of investment and timeline, we suggest that:

1 – ALL of the keywords we’ve already suggested be used to some degree, in a content-generating effort (that’s tracked), to start the testing process. This means creation of some new articles, pages and blog entries.

2 – ALL of the keywords be set up with a PPC (pay-per-click) campaign, to begin testing and sorting results, and zeroing in on (A) those keywords that generate clicks, and (B) those keywords that tend to result in conversions.

3 – The quickest way to understand EXACTLY which keywords are working or not, is to perform ongoing analytics on the web traffic to ALL CUSTOMER XXX online properties, including any new ones set up (as we’ve suggested) specifically for “social media” purposes (like an industry blog, or article directory).

4 – The following keywords should be given significant focus in terms of use and webpage density, because of the reasons we’ve expressed above – and to simply “capture the market” so to speak with respect to these core CUSTOMER XXX offerings:

• keyword1 – use lower case vs. upper case; and where possible closely follow with a descriptor (or the full phrase), to differentiate it from the software company xxx.

• keyword2 – use the plural, and lower case

• keyword3 - plural and lower case

• keyword4 – lower case; this was actually the term that ranked the highest in WordTracker results, with over 500 searches over the last 3 months; though the acronym can be ascribed to MANY other things than “xxx” – therefore, like keyword1, should be closely associated in text with the phrase “xxx” or even “xxx”

• keyword5 – lower case, plural – this is THE keyword to focus on, very industry relevant, AND no one’s currently bidding for this on PPC (vs. the singular “xxx”) – lots of opportunity to win here, plus CUSTOMER doesn’t rank in the top 100 on Google at this time, for this word. It should be fairly easy to improve – especially since the CUSTOMER XXX front page is so lacking with respect to content.

• keyword6 – another very key phrase, ripe for SEM success

Note that WordTracker reports fairly low traffic with respect to these very industry-specific terms, like thermosets – but in the B2B space, most searches are apparently already “qualified” to some extent, i.e. the searcher is not just browsing, but wanting to find sources and detailed information. This means that while the volume of searches may be low, the quality of the prospect and conversion is going to be very high. Ergo, WordTracker results (and tools like this) aren’t especially helpful in the B2B space – manual research, industry knowledge, and above all TESTING is the most important way to get good results.

In summary, focusing on keywords and SEO isn’t really the priority this site should be taking; focusing on building and establishing industry conversation, syndicated material and building a community of interest through SEM (search engine marketing) and SMM (social media marketing) is."


Post a Comment

Subscribe to Post Comments [Atom]

<< Home