How To Find The Best Advertising Keywords For Your Search Engine Marketing Campaign
What can I say, I love keyword research. So much in fact that as a search marketer I make it a point to look closely at the advertisements scattered all across the various search engines. I enjoy seeing other advertisers headlines, especially in ultra-competitive arenas where the average CPC (cost-per-click) for certain keywords reaches $30-$40. But having done this for a while now, I’m willing to bet 90% of companies running PPC (pay-per-click) advertisements across Google/Bing/Yahoo’s search engines aren’t profitable.
This actually reminds me of a quote from one of the greatest advertisers of all time — the one and only Claude Hopkins:
“Advertising is salesmanship. Its principles are the principles of salesmanship. Successes and failures in both lines are due to like causes. Thus every advertising question should be answered by salesman’s standards. Let us emphasize that point. The only purpose of advertising is to make sales. It is profitable or unprofitable according to its actual sales. It is not for general effect. It is not to keep your name before the people. It is not primarily to aid your other salesmen… The difference is only in degree. Advertising is multiplied salesmanship.”
Too many companies think its fine to drum up a couple broad keywords that trigger a slapdash ad copy. And as they look at their data and notice the CPA (cost-per-acquisition) skyrocketing, they continue to move forward with their campaigns thinking they’re receiving some unnamable benefits. I’m going to tell you what PPC advertising is not – it is not for brand building. It’s not to feed your vanity seeing your ad on the first page of Google. PPC advertising is, at its core, what Claude Hopkins called “scientific salesmanship” and to generate the sales you need to make sure your campaign is set up correctly. The first step to doing that is knowing how to do your keyword research.
Finding The Keywords That Make The Most Sense For Your Campaign
How would you put together a search engine advertising campaign for the product below?

We already know there’s a high demand for products in the mobile accessories market. It isn’t difficult to imagine some of the keyword phrases searchers will use. But in order to run a fully optimized campaign, it’s important to analyze related keywords in depth.
If you do a quick search on Google, you’ll notice legions upon legions of advertisers fighting to take ownership of short-tail keywords. Too many companies target ultra-competitive keywords like “iPhone 4 Case” and set the ‘Match type’ to ‘broad,’ allowing the search engines to trigger their advertisements when related terms are used. There are a couple reasons why this method is ineffective:
1) When somebody uses the keyword phrase “iPhone 4 Case” there’s a good chance the person doesn’t really know what she specifically wants. She might be leaning towards a silicon case, a hard plastic case, a slider case, etc. But ultimately, she probably wants to do a little shopping first. Therefore, because the Anne Rossi is a very distinct case, we want to get visitors who are already looking to purchase a case just like it.
2) While the keyword “iPhone 4 Case” might not be as competitive as a keyword like “online car insurance” (Average CPC = $50!), it’s still very competitive, considering the smaller margin of profit for a cell phone case. And when the level of competition rises, the average CPC follows suit. With an average CPC of $1.20, if the Anne Rossi happened to have a margin of profit of $25, then the conversion rate cannot fall short of 5% or else our campaign would be unprofitable.
With these thoughts in mind, we need to begin compiling a long list of words searchers might use to find a product like the Anne Rossi. Using both Google AdWords Keyword Tool, along with Microsoft adCenter’s Keyword Tool, we begin to narrow down search terms to weed out unlikely buyers, and target a group of buyers who are most likely to purchase our case. On Google’s Keyword Tool, when researching “iPhone 4 Case,” we get a list of 787 related keywords with a whole host of useful data, including:
1) Competition
2) Global Monthly Search
3) Local Monthly Search
4) Approximate CPC
5) Local Search Trends
After scrolling down the list, we begin to derive a better sense of more precise search terms and their costs. As these search terms manifest, we’ll find that we can begin adding more creativity to our strategy. For example, we may see a competitor being frequently searched for in conjunction with the search terms we’re targeting (i.e. Coach iPhone 4 Flip Case). If we wanted to be really aggressive and take some of the market share away from our competitor, then we could use this keyword to trigger our ad as well.
Organizing Your Keyword Research
Here’s an example of how one might use related set of keywords and place them in different adgroups under one campaign:
The reason we need to separate different set of keywords within a campaign is because we want the most relevant ad to trigger with a particular search query. So when a person searches for an “iPhone 4S Pouch,” I want my ad to include that term so searchers are more likely to click through.
I’m sure you’ll notice that some of these keywords were seldomly searched for. With a quick search on Google for a keyword like “iPhone 4S Pouch,” you’ll find there are hardly any advertisers there. No one is putting out an aggressive ad campaign to target specifically those who are searching with that query.
The few text ads that are being triggered are mostly companies whose generic ads are showing up because they set their ‘match type’ to ‘broad.’ If this is a market we really wanted to take hold of, we can create a compelling text ad and place a reasonable bid. Because we’ll have a more relevant ad, our CTR is bound to go up. With Google and Bing/Yahoo’s hybrid auction system, as our CTR goes up, the CPC will go down because the search engines favor those whose ads are the most relevant to their users.
Using Negative Keywords To Minimize Wasted Clicks
As you’re setting up your campaign, I would highly recommend that you either use quotations (phrase matches) or brackets (exact matches) around your keywords. With the broad match option, the search engines will automatically trigger your ad with synonyms that might be completely irrelevant to what you’re offering. However, if you choose to use either broad or phrase match, another useful tool would be to include negatives keywords. For example, if I were to have used the phrase match option for “iPhone 4S Pouch,” I would be susceptible to a click-happy, cheapskate searcher who is searching for “Free iPhone 4S Pouch.” By simply adding “-free” to my keyword list, its one unwanted click I have to worry about.
So You Found Your Keywords, Now What?
Once you’ve figured out how to find the keywords that you think will bring you the most qualified clicks as possible, then it’s time to put them to the test by adding them to a live campaign. This is where the rubber hits the road when it comes to seeing what keywords give you the conversions that make the most sense for your business.
At this point a lot of online marketers would just put their campaigns on auto-pilot, but don’t fall into that trap and always be sure to have a good idea of how your campaigns are performing. Optimizing and testing your campaigns is just as important as finding the best keywords, but we’ll leave that topic for another time.
About the Author: Delbert is an internet marketing consultant specializing in pay-per-click and display advertisements and is passionate about helping small businesses understand and develop intelligent internet marketing campaigns. You can connect with him on Twitter @delbert_kim.

{ 3 comments }
Good writeup. The key thing though is to approach this scientifically and actually keep track of the results that you get with different keywords so that you can find out what converts best over the long-term with certainty. Way too many companies are investing in ads, but then don’t really track ROI, which is absurd. Even though Facebook doesn’t compete with search directly (yet) I see a lot of companies using various Facebook ad units and the types of companies listed at http://www.buyfacebookfansreviews.com to test out different types of keywords so companies are experimenting with ads in many ways to see what works best. I think that the key thing though is making sure that you’re keeping track of this data in a scientific and careful manner because way too many companies are not keeping track of these results or doing careful tests to see which of their keywords works best in certain contexts.
Thanks for your comment Will. I completely agree that a scientific salesmanship is needed to optimize different keywords and ad copy to its utmost. When that perfect pair of keyword and ad copy is found and the landing page is optimized to make the sale once prospective customer arrive, there’s aren’t too many business models on the web that is more beautiful.
I do like buying clicks on Search, but I must say that I’m not a fan of the website you listed. I just think purchasing Facebook fans is a hollow way to build your brand.
Another tip for generating negative keywords: obviously set some negative keywords when you first launch your campaign (for example, “free” was a no-brainer for us…I have no desire to pay for clicks from people looking for free software), but also monitor the search terms that are generating paid clicks on an ongoing basis. Our company uses Clicky to monitor web traffic, and Clicky very clearly shows which visits came from paid search, along with the search term that the visitor used. When I see search terms that were a bad fit (e.g., “make a vinyl banner”…in this case “vinyl” is a give-away that the click was a waste of money), I try to dissect them and add negative keywords to AdWords to prevent our ads from triggering for that search term (or anything similar) ever again. And, of course, you have to use Google’s keyword match types wisely so as to not filter out good traffic along with the bad.
Comments on this entry are closed.