Selecting Keywords for SEO: A Quick Guide for PR and Social Media Pros

Shannon Paul’s had a post yesterday that included very good advice for PR pros who want to plunge into the social media world (make sure you look at the presentation she’s embedded in the post).  Shannon suggests that PR pros need to start thinking about how they can make their content searchable and sharable in order to make the leap.  Kudos to Shannon for raising an issue that the clients of PR agencies have been demanding – make it easy to find the information – focus on keywords, SEO and links.

Given that the intersection of social media, PR and SEO is a topic that’s near and dear to our hearts here at BuzzStream, I thought I’d expand on one of the topics in Shannon’s presentation – keyword selection.  Picking keywords is incredibly important, and not just for press release optimization…do it right and it will help all of your marketing activities.

For our SEO-oriented audience, most of this will be fairly basic.  For those of you in PR that are new to this, I’m hoping it will give you some good ideas about how you can more effectively identify keywords, and do it in a fast, inexpensive fashion.  There’s no one right way to select keywords, but we like the approach I’m going to describe because it helps you identify keywords that are closely aligned to the terms your customer uses to shop for or to find information about products in your market (as opposed to simply finding keywords based on things like overall keyword popularity).

Keyword selection can feel pretty daunting when you’re just getting started, but it’s not as tough as it seems.  Here’s how we do it at BuzzStream.

Don’t START with Google’s Keyword Suggestion Tool!

Note that I didn’t say “don’t use the keyword suggestion tool.”  It’s valuable as a supplemental tool, but in my opinion there are a lot of reasons not to rely on it as your starting point.  The problems are similar in many ways to the problems with relying on shotgun blast media pitches for your media and blogger outreach efforts…it’s broad-based, but much of what you get is irrelevant.  Additionally, it doesn’t help you identify the long-tail search opportunities, which have a ton of potential value.  Instead, you need to start by trying to put yourself in the customer’s shoes (if you’ve developed personas and a positioning statement for the company, it’ll be even easier).  In order to do this, the first thing we do is brainstorm on the following topics …for each, I’ve included some of the more general terms we’ve identified for BuzzStream’s customer to serve as examples:

  1. Who is the product for? – e.g., small business, SMBs, DIY
  2. What type/category? – e.g., marketing, word-of-mouth, SEO, public relations
  3. What is it? – e.g., software, service, tools
  4. Verbs/adjectives? – e.g., improve enhance, better
  5. What does it affect? – PageRank, publicity, lead generation

For each of these, start with the most general terms and progressively drill-down.  So, for example, you might have “marketing” as the most general term for “category,” and from there you might drill all the way down to something as specific as “microPR.”  The more general terms will have much more traffic, but they’re harder to rank on and they don’t convert as well.  It’s the exact opposite for the more specific terms, which is what makes them so valuable.

Once you’re done, you’ll end up with a bunch of keywords in each of the five categories.  Then you start putting the terms together – e.g., “small business marketing software,” and “tools to improve search performance.”  You can do this in Excel, so that you don’t have to manually create the combinations.  You’ll need to eyeball the combinations and remove the ones that don’t make sense…you don’t have to spend a ton of time doing this because the bad ones will mostly be thrown out when you test your keywords (I’ll cover this in a minute).

Check out the competition

You can supplement the concept-oriented keywords you created by looking at your competition to see what they’re doing.  There are lots of tools to help you see what others are bidding on and to see their ads.  This is valuable because you get to see the language they use in their ads…it also helps you identify competitors that you weren’t aware of.  Some of the tools to look at include adgooroo, spyfu and keycompete.  All of these tools include a free trial period.

Competitive keyword searching still won’t tell you which terms are working and not working though.  For that, you need to test.

Test, test, test!

Once you’ve generated your keywords combinations, you can test them with an Adwords campaign.  Setting up an adwords campaign is easy to do and it’s inexpensive.  You can take a very large list of keywords (thousands) and get a good idea of what your customer really care about for less than a $1,000.  The information you’ll get back is incredibly useful because not only do you find out what people are clicking on, you can determine what converts into blog subscriptions, email signups, leads, revenue, etc.

Other resources

This is really just the tip of the iceberg, and there are a ton of good resources if you want to dig in deep into keyword research and selection.  My favorite is Search Engine Guide’s series on keyword research, selection and organization.  Aaron Wall has great training information on keyword selection as well.

If there are specific areas of keyword selection you’d like us to drill into in future posts, let us know.

One other thing – keyword selection is as much art as science, so feel free to jump in here…PR and social media pros – what’s working well for you when selecting keywords?


  • Jeremy

    Hey Pam, if you’re working with a large set of prospective keywords, I wouldn’t expect that the strategy of writing a general post to test rankability would work except for the most obscure and least-searched terms.

    What you can do is screen your list in Hubspot or Wordtracker by competitiveness. Those tools have a competitiveness score that can tell you which ones are just too competitive for you.

    The long-tail is a no-brainer because all you really have to do is build content and you’ll rank.

    It’s the magic middle that gets you– the terms where there is some competition, but the big unknown is really how valuable they are likely to be. That’s where you’d want to use the approach outlined above to calculate your conversions by keywords. Then you should just aim at ranking for the terms with the highest value and pay less attention to their competitiveness. Like you suggest, there’s really no way to know how you’ll do until you try.

    So overall this strategy says: don’t waste time on the terms that are too competitive or the terms that are least valuable to /you/ as measured by a PPC campaign with conversion testing.

  • Great advice Paul. Another area that interests me is understanding how to balance your selection between high volume/highly competitive terms and long tail terms.

    We’ve had a pretty easy time achieving results on Google for our long tail terms, but have found that we could invest too much time without capturing much value on some highly competitive terms.

    The question that always plagued me was: How big is too big when you are working with limited resources? Should I go for big results on lower volume keywords or go for the gold and climb to the top on a high volume terms? As a type A, you can imagine my preference but my SEO guy was not always so thrilled with that notion because it takes so long to know whether you will see results.

    You can continue struggling for that term for weeks or even months before figuring out you didn’t have a chance because Google didn’t/doesn’t?? crawl typical webpages for weeks. Not sure that’s still true? One of the tricks we used to test our possibilities was to post content optimized for a broad keyword on a social media site or our blog (those are cached in just days) to see how we fared. With that we had an idea of the results we could attain with a little effort. Not sure if this is a good idea or if it is even current as SEO changes so rapidly. Thoughts?