Paid Search Marketing
The question “are long tail keywords dead for paid search” is fundamentally the wrong question.
There are a few things at play here:
The long tail of search is alive and well.
The long tail of search queries is apparent in both paid and natural search marketing.
The tail of keywords in paid search has actually become pretty short.
If you’re confused, it’s probably because the above three points assume a seldom-made distinction:
Keywords are not search queries.
A keyword is a word or phrase you’ve decided to target in your marketing efforts. A search query is the actual thing a searcher types in before they see your listing or ad and navigate to your site.
It’s an important distinction when talking about the long tail and pay-per click marketing, because while the existence of a long tail of search queries is essentially indisputable, looking at the keywords in your PPC accounts may leave you wondering what happened to this mythical “long tail”.
We’ll talk a bit about both why this is and what to do about it in the coming paragraphs.
Whether You Acknowledge It Or Not, The Long Tail is Real
There is little doubt that searchers’ behavior creates a long tail of search queries. In our post on profiting from long tail keywords, we noted the following statistics:
What Google describes as long tail advertisers make up half their revenue
Searchers are using longer queries
So if there’s no dispute that a long tail of search exists, why are industry thought leaders saying that the long tail of keywords is dead?
The Keywords in Your PPC Account Aren't the Only Way People Reached Your Site
The problem here lies in the disconnect between the things you seem to be bidding on and the things you actually buy:
Here we see a fairly standard example of the way keywords are mapped to the different AdWords matching types. The thing to note here is that the keyword Adidas shoes is matching to the query nike slippers. This sort of aggressive association isn’t an anomaly. So, your Adidas shoes keyword may be sopping up search queries like:
- Adidas slippers
- Nike shoes
- Adidas basketball shoes
- Basketball shoes
- Addidas shoes (misspell)
- And more.
Naturally, this leads to top-heavy paid search accounts. A short list of keywords will see a handful of the broadest variations getting the greatest amount of traffic and conversions. And, in turn, those keywords get more attention in the form of bid manipulation and account structure.
So the question becomes: if you’re running a paid search campaign, why should you care?
Better Information -> Better Actions -> Better Results
Analyzing a keyword leaves you with insufficient insight. The same keyword may be matching against:
- Queries that convert exceptionally well
- Queries that could be converting better with some help
- Queries that are completely irrelevant.
Then, it treats them all exactly the same. If you were selling Adidas shoes, would you place the exact same value on the following searches:
- Buy Adidas shoes
- Adidas basketball shoes
- Nike slippers
Obviously the three searchers have very different values to your business. What we’re proposing is that you should have insight into not only keyword-level traffic and conversion data, but also query-level traffic and conversion information.
While it seems obvious that higher PPC resolution is a good thing, the trick is knowing how to leverage the data.
The Elements You Can Effect with Search Query Data
Once you start to gain insight into the search queries driving traffic via your paid search campaign, you can use that information to impact:
- Keyword Selection & Keyword Analysis: Looking more closely at search queries allows you to see which specific queries are performing, so that you can integrate those into your account.
- Bidding: Instead of adjusting bids on broad, umbrella keywords, you can react to specific queries with more granular bidding strategies.
- Negative Keywords: With visibility into the queries a keyword is shown against, you can effectively weed out “garbage” or irrelevant clicks.
- Match Types: With increased visibility into how individual queries perform, you can implement a sophisticated matching strategy aimed at capturing and bidding traffic based on query-level performance.
Basically, with visibility into your long tail of search queries, you can drastically alter the way you perform every paid search activity. You can bid keywords differently, pick different keywords, even assign them different match types based on this extra information.
But what about when “extra information” becomes “too much” information?
Cutting Through the Fog of Long Tail Search Queries
The problem here lies in the actual analysis and integration of this more granular information.
In other words: how do you keep your account from being overrun by low-volume, low-return keywords?
The Overwhelming Power of Strong PPC Campaign Organization
The most important thing in integrating new keywords into your campaign is where you put them. If you don’t have a very strong, well-organized campaign introducing keywords becomes a very messy (and financially inefficient) process.
Creating intelligent keyword clusters makes introducing successful queries, bidding, and adjusting match types a much simpler process. What I mean by an “intelligent keyword cluster” is basically just that you should create tight, semantically related keyword groups:
Here we see a keyword taxonomy around the word “Adidas”. By creating an intelligent grouping structure that moves from broad to specific, we’re able to enjoy the best of both worlds where the long term of search is concerned.
Within our Adidas shoes group, we can take a high (Ad Group) level view of all of the Adidas shoe terms driving traffic. The powerful thing about tracking to the query level and then creating close segmentations is that we can look at varying levels of performance. We can take a birds-eye peak at the group level to see how Adidas shoes is doing:
Or we can get a more granular peek behind the curtain at how each Adidas shoe query is holding up:
The key is by creating tight semantic clusters so that we retain the choice to zoom in or out. Seeing that the misspell of Adidas here has driven one conversion in two visits could be extremely valuable information. If a keyword has insufficient data surrounding it, we can jump back up to the group level. If we want to take a deeper dive and gain more information, that option is available as well.
From here, we can identify:
- The queries that perform - Raise their bids or find similar terms and create a brand new segmentation (to which you can specially craft ad text and finely tune bids).
- The queries that don’t – We can either try a different matching option and bid, or nuke them altogether if they ‘re truly irrelevant.
- The truth about keywords – Is it the keyword you’re bidding on that’s doing well, or a query it’s being matched against? What if you bid on the specific queries driving traffic, filtered out the irrelevant clicks, and started to bid that keyword you thought was so successful based on the way it’s actually performing?
So that we can positively impact the bottom line of our PPC accounts with better-informed keyword selection, bid strategy, and match-type implementation.
So the Long Tail is...Alive, Then?
The long tail of search queries is very much alive. The question isn’t “is there a long tail of PPC keywords” or even “should there be”. Ultimately, the question is “do you know enough about your own long tail of search queries to execute on your PPC keywords”?