If you are reading this article you have either made the decision to significantly restructure your AdWords account or you are seriously considering it.
Chances are, you are doing this because your account suffers from what I call the broad match effect. The purpose of this article is to prepare you for the journey we will be on as we transition from your current account structure to the new one.
Making any change to your AdWords account involves some degree of risk. AdWords is an extremely complex and dynamic system. If you allow me to restructure your account, there will be certain risks, but the risk is minimized because we know:
- what the problem is,
- what to do about it,
- what the result will be and
- what to expect during the transition.
I want to elaborate on this because there will be short-term setbacks and long-term ramifications. You may see some alarming performance data during the transition. Some of this performance data is real and represents actual short-term and long-term higher costs and some of it is not as real, just misleading.
The real part comes two factors; higher CPC associated with more restrictive keyword matching options and from the fact that we will be reestablishing quality scores for keywords and ad copy. Yes, ad copy has a quality score, you just can’t see it like you can for keywords, but it is real and it does have an impact on performance. Finding the balance of ad quality and optimal ad CTR is tricky business and can only be accomplished through testing. In addition, I recommend giving your Max CPC bids a boost in the early going to establish a higher CTR. As you know, the biggest component of keyword quality score is CTR.
Here is a graphic I created to illustrate how keyword matching options effect your CPC, exposure and quality of visitors.
This is a an alarming reality for some advertisers and you should be prepared to spend 50-500% more CPC when you make this Major Chance to your campaign structure!
The part which is not so obvious is your cost per conversion. Because conversion tracking uses a 30-day cookie, cost/conv data for any given keyword on any given day will change based on newly registered conversions. That’s because conversions and transactions get attributed to the date of the click (on your ad), not the date the conversion occurred.
Here is something for you to think about. If you want to know the effect a change you make today has on cost/conv, you need to wait at least sixty days and then look back thirty days. Otherwise you will be dealing with incomplete and misleading data. You need to resist evaluating the new campaign based on short-term data. Also keep in mind that the number of impressions and clicks can vary considerably from day-to-day, week-to-week and month-to-month. You must be willing to have the necessary confidence, commitment and patience or you should not be doing this.
Depending on your market, conversion tracking data can increase many fold over the life of the cookie, as visitors return to your site (from non-AdWords links) and convert. I make the distinction, “from non-AdWords links”, such as a bookmark, organic listing or website referrals, because AdWords attributes the conversion to the last-clicked ad, keyword and search term, even if the keyword or ad is paused.
I make the point about paused keywords because you will continue to register conversions in your account from keywords in your old campaigns, which may be paused, and you probably won’t be looking at those keywords when we launch the new campaign(s).
The broad-match wild card effect – With all the down-side of using broad-matched keywords, there is one positive aspect worth mentioning, and it can be significant.
The fact is that some search queries can only trigger ads using broad-matched keywords. Unless you have spent considerable time pouring through server logs and search term reports, you probably don’t realize just how many possible unique search queries there are for your broad-matched keywords. It’s essentially unlimited. Some of them are great and some are dogs. The dogs are why you have been racking up all those relatively poor quality impressions and clicks. But notice just how many of those strange ones produced conversions! Chances are that those strange search terms will not trigger your ad with anything other than a broad-matched keyword. If you tried to target that same search query with a more targeted keyword match type, you would almost certainly be given a status of Low search volume.
Here is something else to consider. When you use broad-matched keywords, you become eligible for a very large number of ad auctions. At the time of the ad auction, only a fraction of those eligible, actually get displayed. There are so many dynamic factors at play, that it becomes unpredictable, especially when you have an unusual search query. I call this the broad-match wildcard effect. While you might be very happy that your ad was displayed for a particular search query, it will be impossible to insure that it will be displayed the next time that search query is performed.
While we will try to create keywords and matching options that display your ad for as many relevant search queries as possible, there is only so much we can do. We don’t have complete control over the AdWords system and it isn’t 100% predictable. Therefore, if we use more restrictive keyword match types, you must be willing to accept the fact that we will not be able to display your ad for every conceivable search query. If your logs indicate that you have received a significant number of conversions from search queries that we cannot reliably link to a keyword in the new design, we should not implement a completely new design. A partial redesign, focusing on a segment of the account or campaign may be in order.
The importance of fine-tuning – Most advertisers can’t appreciate the importance of or the amount of time required to fine-tune a campaign once it’s been launched. No one ever get’s everything right when the campaign is first launched.
When the initial campaign is launched, default bid prices are used, which amount to little more than a guess as to what the bid price should be in order to receive the desired ad ranking for each keyword. It’s simply not practical to research bid prices for a large number of keywords prior to launch. In the beginning it will be necessary for me to monitor and make adjustments at least once a day and then it tails off over time. The amount of time this requires depends on the number of keywords, ad groups and campaigns in your account.
Each keyword is unique and is actually a “market” unto itself. Therefore, until the keyword is actually searched on and your ads receive “clicks”, you can’t be sure where your ad will appear in the ranking or how much you need to bid to obtain the rank that is appropriate. Some keywords are simply much more valuable to you than others. It is an iterative process and the two most important factors are 1) how much traffic your campaign receives and 2) how much time I have to work on it. Some of my clients have made a serious mistake by not allowing me to perform the fine-tuning process once the new campaign is launched or they underestimate the amount of work required to perform this important task.
Before we begin the transition, I recommend that you implement Google Analytics. It offers data you won’t see in AdWords, such as bounce rate, number of pages per visit, % new visitors and average time on site. These are all valuable pieces of data that help us establish relative values for specific keywords, ad copy and web pages. In addition, you can set up “goals”, which are like conversions. But unlike the AdWords 30-day cookie, the Google Analytics cookie can last up to two years! This can help give us a more accurate cost/conv.
So let’s review why you might want to incur the additional expense of redesigning your AdWords account and seemingly subject yourself to the risk.
The characteristics I describe in the broad match effect, implies you don’t have the desired control over what’s going on in your AdWords account. You aren’t able to control what ad a user sees for any given search query, your ads are being shown for irrelevant searches (unwanted impressions), you are paying for a lot of poor quality visitors and you are spending a lot of time making changes to your account without any real confidence you know what you are doing. You’re doing stuff like; ad copy changes, new keywords, experimenting with keyword matching options, adding a lot of negative keywords and changing bid prices. Probably not the best use of your time.
The answer is, you will get:
- Better search query to ad copy relevance
- Higher CTR
- Higher quality scores
- Lower CPC
- Lower cost/conv
- Fewer unwanted impressions
- Higher CTR
- Lower CPC
- Lower cost/conv
- Fewer unwanted clicks
- Lower ad spend
- Lower cost/conv
- Easier on-going account maintenance
- Lower costs
- Less of your time
- Better ROI
- The confidence in knowing what’s actually happening
- Frees you up to focus on other things
- The confidence in knowing you are in control
- Freedom to try new things
- A happier life 🙂
One final word of caution. You have probably come to the realization that the new account design I discussed is pretty involved. Well, you’re right. Therefore, if you or anyone else make changes to the new campaign, chances are they will do more harm than good. It would be like hiring someone to tune your baby grand piano and then making a few minor adjustments on your own! Not a good idea.