How to Capture & Control Your PPC Keywords to Achieve a Better Account Structure
When we think of PPC keywords, we think of it in terms of capture versus control. On the search query side, you’re looking to find where you have profitable volume; and this is known as capturing data.
Once you have high volume, profitable words, then you need to control the keyword in terms of monitoring and bidding.
By separating out campaigns between capture and control, we can then control budgets and workflow quite easily so that we keep our clients profitable, and yet continue to grow their accounts within the constraints of our methodology.
For PPC Associates, internally we call these alpha (control) and beta (capture) campaigns.
The Alpha Beta Structure
At the core of the Alpha Beta structure are two separate campaigns (or sets of campaigns) – “Alpha” and “Beta.” Ultimately, Alphas are campaigns that have the primary goal of ‘query optimization,’ while Beta campaigns focus on the goal of ‘new query capture ’ (subject to profitability constraints). All keywords in Alpha are proven to be high-volume queries (exact match keywords), whereas all keywords in Beta are still in testing mode (non-exact match keywords, trying to find new high-volume queries).
Let’s assume that you just opened up an account in AdWords. Because you have no history of success on a keyword or query basis, you start with one campaign – a Beta. The structure of your campaign should consist of a series of highly targeted ad groups. Each ad group should have targeted ad text that speaks to the intent of the keywords in that ad group, and preferably you have landing pages that also relate specifically to the keywords.
Query Capture – Creating a Beta Campaign
The keywords that you have created on broad match modified in your Beta campaign are your “bait” – you use these to get Google’s algorithm to match you on related queries. Every time Google matches you, it’s like a fish nibbling on a lure – hence the “bait” term. In a few days, you’ll start to see a lot of different queries in your account – some converting, and some that don’t. You need to set a threshold of clicks, cost, and conversion to assess the value of these queries. For example, you might set a cost per acquisition (CPA) objective of $15 and decide that you need at least three conversions at $15 or less to conclude that a query is high enough volume for you to spend additional time on ‘query optimization.’
On a regular basis, you’ll run a search query report and assess all queries based on the query’s metrics.
The next step is to take this data and move keywords into the appropriate parts of the account. This is where the advantages of Alpha Beta will become apparent!
Targeting the High-Volume Winners – Creating the Alpha Campaign(s)
Your Alpha campaign is for proven high-volume queries, where high volume is defined by the account manager and how much how time he/she has to invest in query optimization.
In your Beta campaign, you set keywords to broad match modified, or some other non-exact match type.
Again, in the Alpha campaign, keywords (which are based on specific queries) are set to exact match. Moreover, each keyword is put into its own ad group. Internally, we call this “skagging” because you create Single Keyword Ad Groups (SKAGs). Note: since it’s really about queries, we should call is SQAGs, but SKAGs has a better ring to it.
Putting winning queries on exact match in their own ad groups has several huge benefits. First, because you are only using exact match throughout your Alpha campaign, you should only have winning, profitable, precisely crafted high-volume keywords in your Alpha campaign (or other high-volume queries that perform near enough to the goal that they should not be negatives), and there is no way for Google to match you to ‘losers.’ Second, the SKAG process enables you to create highly targeted ad text and landing pages. For example, for the keyword “Carnegie Hall Hotels,” you can create an ad that mentions “Carnegie Hall Hotels” a couple of times. Because Google will bold any instance of the exact keyword in your ad text, your ad appears more relevant to a searcher and is likely to drive higher click-through rates and lower CPCs. You can also create a custom landing page just for that keyword or choose the most targeted landing page possible. This should result in higher conversion. Combine higher CTRs, lower CPCs, higher conversion rate, and no poor matches, and you should see immediate improvement to your ROI.
Forcing Google to Send Traffic to ‘Alphas’ – Controlling Query-to-Keyword Mappings
Now that you have great Alpha keywords with targeted ad text and landing pages, as well as very precise bids, you need to make sure Google does not serve your Alpha queries in your Beta campaign. This ensures proper keyword-to-query mappings, so that all the data for Alpha queries is housed within the Alpha campaign.
It turns out that there is often a difference in how Google says their system serves keywords and how Google actually serves keywords. According to Google’s AdWords help center, if a keyword in your account is on exact match and exactly matches a user’s query, that exact match keyword will be served, even if you have another broad match (or broad match modified, or phrase match) keyword that could potentially be served. Here’s Google’s official explanation of their prioritization process.
In truth, Google’s system will sometimes serve a broad match keyword, even if you have that query as an exact match keyword. This can be harmful to your account in two main ways: first, it may serve ad text and a landing page that is not as targeted as those in your Alpha campaign; second, it may actually cost you more per click on broad match than on exact match. Typically, Google takes every opportunity to map a query to the keyword with the highest ad rank (see the list of exceptions in the above referenced Google help article), which usually equates to sending an impression to the keyword with the highest bid. Another benefit of forcing traffic to Alphas is that it will ensure all data for one query is sent to the same place and, just as important, NOT sent to some broad match variation (which could result in the broad match version looking better than it really is, because it contains data of a ‘known high-volume query’).
The solution to this problem is quite simple: every time you add a query as an exact match keyword in your Alpha campaign, add it as an exact match negative keyword to your Beta campaign.
Don’t Forget About Optimizing ‘Query Capture’ Beta Campaigns
Creating a great Alpha campaign without simultaneously policing your Beta campaign is a recipe for disaster. So, in order to ensure an entire account (not just Alphas) has good performance, it’s important to continue with optimization efforts within Beta campaigns. In Beta, you’ll still be able to bid with some degree of precision, find the best performing ad for the aggregate of your queries in an ad group, and pick landing pages that perform best on average for the ad group. So, there are a lot optimization levers still available, but they are not as precise as ‘query optimization’ Alpha campaign efforts. In Beta, you’re catering to the mean, while Alpha queries are all independently optimized. White it’s a bit hard to stomach that Betas ‘cater to the mean,’ it’s also important to remember that they have relatively low volume. So, mathematical precision at the query level isn’t an option, thus the best available data is the ‘clustered’ data created by the original tightly themed ad groups.
For a more comprehensive description of the Alpha Beta Account Structure and its benefits, visit the PPC Associates download page.
Note from Brad: Mike Nelson is the next guest on Marketing Nirvana and will go over the control and capture methodology as well as many other insights into organization on the show. The show will air June 4th at 9am PST / 12 EST at Webmaster Radio.
Opinions expressed in the article are those of the guest author and not necessarily Certified Knowledge. If you would like to write for Certified Knowledge, please let us know.
Nice post. I do something similar, and on a smaller scale, for select Ad Groups. I think I might borrow your Alpha & Beta terms;)
For man y of my accounts I’m not in favour of this solution, because a) it creates a lot of additional campaigns to be managed and created (having to split keyword search volume across Alpha & Beta), b) individual keywords often don’t get enough conversion traffic to determine it’s effectiveness (especially in niche markets, for smaller clients or local advertising).
I’d rather segment the campaigns based on volume (impressions, clicks and/or conversions), using various match types in an account. You can avoid the mentioned Alpha-Beta competition by downbidding your broad match keywords in your ad group. Usually, we see that Google prefers the exact match anyway.
This technique has the following advantages: a) all keywords profit from ad optimization and relevancy, b) ad AB-testing speeds up considerably, c) a focus on impression volume boosts and guards your accounts’ quality score, d) using real keywords provides insight into related queries and shows which keywords really to target!
Some notes: be careful with broad match: check in the real keywords what it’s matched to (I LOVE +modified +broad +match!). Use negatives across ad groups to avoid internal competition: a clean querie flow provides more predictable results for bidmanagement and allows for faster ad testing.
Any comments on this strategy are very welcome!
@olafpil I personally don’t believe there is a ‘one strategy fits all’ solution. I like the concept of this idea (sometimes you might not implement it 100%; but the concept of ‘search and control’) is really appealing no matter if you use it by campaign, or control it by negatives across ad groups within a campaign.
Some account (as you noted) are local with small budgets; and you struggle with volume and will use more modified broad match (especially in niche industries) to try and find profitable volume. If the companies aren’t using call tracking; then you also end up dealing with really thin data.
You have other large companies that don’t have many landing pages (lead gen); but might spend 7 figures a month, and this is a great strategy for those accounts.
Some sites have budgets by division within the company; so often campaign management is also budget management by division.
Then you have your solid mid-sized accounts. They often spend $10k-$100k/month (could be more/less). Can be a bit nimble with landing pages (assuming is doesn’t have to go through legal) and trying out new features, tests, etc is something they are often better at than small or huge accounts. Often these companies can be flexible in strategy or try out a few without too many issues.
So, in the end – I think you need to find a strategy that you feel comfortable with and fits the account’s data and time management.
Hi Brad, thanks for your reply!
You’re right. In the end, it’s up to the marketeers to determine the strategy to use. Unfortunately, I don’t have any 7-figures-a-month campaign to test the Alpha-Beta strategy. 🙂
It’s always good to read about different strategies on your website!
Hello! Thanks for this post, interesting approach.
I like the fact I’ve been working on a similar technique, just as Olafpijl says I avoid making more campaigns (we also have quite enough), but I do the cross negative action between adgroups inside of a campaign, splitting terms as much as possible and going as far to the long tail as Adwords allows me. When traffic is low, then I fully do what Olaf suggests, I put other matches and BMMs together in the same adgroup (highly relevant) and decrease bid only for the BMM. Also I’m talking about 5 figures a month campaigns or smaller.
I would also add that for me this is important not only to spot the Alpha terms on the Beta adgroups, but also it helps dividing the traffic and getting a clearer read of conversion rates on adgroup level.
One downside of these techniques is when seasonality changes, some of the Keywords stop receiving traffic (low search volume) and then I have to be deleting the negatives back & forth of the “Beta” adgroups!
regards
Really good post. I’ve seen and used variations of this technique but you really have it nailed in terms of level of detail. Thanks.
Great article, I do the same thing for the majority of my clients, however I tend to do it within the same campaign.
so I create individual adgroups with exact match terms but within the same campaign.
If campaign budget was a limiting factor I would move the high converting exact match terms into their own dedicated campaign.
Great post
I like your alpha and beta analogy.
I often use a technique I call ‘Broad Match Generation’: http://www.calculatemarketing.com/blog/techniques/google-adwords-broad-match-generator/.
By keeping your broad-matched keywords in a separate campaign from your exact (and phrase) keywords, and adding exact (and phrase) negatives of all positive keywords to your broad match campaign, your broad match campaign will not canibalise your exact and phrase match queries, as any search query matched to your broad match campaign could not have matched to a keyword in an alpha campaign.
It’s then a case of mining your beta search queries, and if relevant, as as positives exact (and phrase) match to your alpha campaign and negatives to your beta campaign, and if not relevant, add as negatives to all campaigns.
Brad, I believe I will use the Command & Control name from now on, with a reference to Alpha Beta, better describes the overall purpose of this account structuring… your exact campaigns have a distinct impression share (command) boost, and you really have an excellent separation of queries (control), keeping your exact kws separate so as to be able to optimize for “query optimization”. I presented last week at SMX Israel: http://www.slideshare.net/PPCPROZ/alpha-beta-ppc-setup-optimization-30587108 Be sure to play the videos.
https://docs.google.com/presentation/d/1yvc58ELMEzVMOc4Gv3baN0C8fXMQBv3URKRBa0qoG2g/pub?start=false&loop=false&delayms=3000 videos clickable here.