Putting the Short Tail to Work
If you're new here, you may want to subscribe to our RSS feed. Thanks for visiting!
When optimizing an existing campaign, I commonly encounter a setup consisting of primarily short tail keywords to get potential customers to a client’s site. Less frequently, but often enough I come across campaigns that employ a strategy of relying solely on longer tail keywords while completely ignoring the short tail.
I recommend testing a combined strategy that covers a comprehensive range of long and short tail keywords. This often works well to acquire incremental traffic and business. In addition, shorter tail keywords can have the effect of improving long tail performance and can get potential customers to your site before they even hear of a competitor who isn’t employing this strategy.
What are Short Tail Keywords?
Short tail keywords are keywords that are broader or more generic in nature, the opposite being their long tail cousins which are more targeted and specific. For example, a chain of luxury hotels operating in various locations across the globe, might employ the use of such long tail keywords as “luxury hotel in Sydney Australia” for their Sydney location but might also want to use such short tail keywords as “Sydney hotels” or even just “Sydney” to cast a wider net. This strategy sounds dangerous and risky but it doesn’t have to be.
How to Implement and Test
A short tail strategy does have its risks and should be implemented in a controlled environment. I recommend placing these keywords in their own campaign with a separate budget so that they don’t negatively interfere with your main campaign. Starting with a lower bid price and position to minimize the cost of and exposure to less targeted traffic would be advisable. At the same time you want to make sure to have a good list of negative keywords added to the campaign if you plan on using broad and phrase matched keywords.
Once you’ve done this, you’ll want to keep a close eye on the campaign especially at the beginning so you can weed out any inefficiency quickly. Such inefficiencies include keywords that are driving a disproportionate amount of traffic and eating up most of the budget which doesn’t allow for proper testing on all keywords in the campaign. At the same time you’ll want to measure the direct effect on your long tail keywords. Often you’ll find that the presence of short tail keywords increases the performance of the long tail ones directly. You’ll be able to see this depending on the traffic volume your campaigns capture. You should conduct further testing by isolating higher traffic keywords to see what impact they directly have on your long tail keywords.
Why This Strategy Is Worth Testing
There is a bit of psychology and common sense involved in the reason why a combined short tail strategy works. Most people begin their search using broader terms and then refine their search queries when the results generated are too untargeted. So there is a good chance that they will click on your ad if you are offering them what they are looking for before they have to refine their query and search again. Even if they don’t click on your ad after searching on a broader query, you may benefit from the additional impressions and exposure at this stage of their research, causing a stronger association with your brand on subsequent searches and an increased likelihood of getting a potential client to click through to your site.
Conclusion
As with any pay-per-click strategy, testing and experimentation is a key factor to success. So be sure to simultaneously test not only a variety of short tail keywords but different match types as well as highly targeted ads that work to filter out untargeted traffic. From experience I’ve seen that the type of industry that you’re in and the type of product or service that you offer can largely influence the success or failure of such a strategy so results will definitely vary.
Tried this strategy and it worked/failed for you? We’d love to hear about your experience!
Sphinn it! or
Hi Xurxo,
Quite interesting. I would like you to develop a little (in another post) on how that Short/Long Tail strategy plays out with keywords being more expensive in the Short, and usually having a smaller conversion rate, whereas Long Tail phrases would be cheaper with a better conversion rate, since the visitor has refined his/her search and is more advanced in the purchase cycle. Well, that’s what I would think (I may be wrong here).
So, I guess determining the equilibrium of both Short/Long Tails for maximum revenue at a minimum cost can be very hard. Any thoughts?
This looks a most useful approach, Xurxo. I thought it deserved a wider audience, so I Sphunn and Stumbled it. :)
@Jacques:
You’ve raised some interesting questions. In regards to keyword price along the tail, this really depends on the market you’re advertising in. In most cases, the total cost of a short tail keyword may be more expensive due to the higher search volume. However, the average CPC of a long tail keyword may be much higher but with lower search volume. Taking “book a luxury hotels in Sydney” as an example: if the searcher is more advanced in the purchase cycle vs the query “Sydney hotels” and may convert better, this most likely will be reflected by the competitiveness on this keyword resulting in a higher CPC. Overall, this longer tail keyword may receive less search volume and less overall costs but the average CPC could be much higher since the value of the keyword is higher. Since it’s critical to achieve a desirable cost per acquisition (CPA) and since short tail keywords generally can be of lower value from a strict ROI standpoint, the strategy remains to experiment and test in a controlled manner and make ongoing adjustments.
As for the equilibrium point, short answer: Yes, it’s not easy to pinpoint! There are many variables that are in a constant state of flux. Ex. Click costs, seasonality and competitiveness of the market just to name a few. The important thing is to test, measure, and improve continuously in a structured and as controlled process as possible to maximize desired results.
@Barry:
Thanks for the stumble and the sphinn!