|
 Have you experimented with Google AdWords Automatic Matching? Google bills it as a way to bid on relevant keywords that you may have overlooked. We have a different view. It's a good supplementary tool if you've already done solid keyword-discovery research, but it's a potential hazard if you haven't.
How Automatic Matching Works
Automatic matching is an optional feature that Google first began beta-testing last spring. If you enable it for a specific campaign, Google will analyze your ads, keywords and landing pages to deduce other keywords that might be relevant. When users search for possibly relevant terms that you haven't bid on, Google places a bid and displays your ad.
According to Google, there are controls in place to make sure you don't exceed your budget and that automatic matching doesn't interfere with your normal bids.
So what could go wrong? If I'm relatively new to search marketing and I'm not quite sure which keywords to bid on, why not let Google decide for me? After all, they're Google, the kings of search.
Um … well … if you're new to search marketing and haven't quite fleshed out your keyword list, your chances of getting irrelevant matches are much higher than if you're experienced and have a strong list. And that means you could end up spending a huge chunk of your budget on traffic that's mediocre at best.
Our Automatic-Matching Road Test
Lyris recently turned automatic matching on for two campaigns and let it run for two months. We had already done extensive keyword-discovery research and built out a list of thousands of long-tail keywords. (Read Generating Keywords from the Inside Out for ideas on how to do this.)
What we found is that relative to the overall performance of the two campaigns in question, automatic matching:
-
Brought in less than 1 percent of our total traffic
-
Had a click-through rate five times higher than our overall average, but a conversion rate that was only half as good
-
Was low cost, with a cost-per-click that was about half of the average.
Nothing earth-shattering or game-changing, for better or for worse … until we looked at the individual search queries that triggered the automatic matching.
We did indeed find some relevant keyword phrases that we hadn't considered. They tended to be low-volume search terms, toward the far end of the long-tail spectrum.
But about 60 percent of the presumably relevant keywords weren't relevant at all. For example, one of our ads was about "Free HTML email templates" and automatic matching bid on terms like "invoice templates" and "affidavit templates" that had nothing to do with email.
Implications for Marketers
Based on our road test, we believe that experienced search marketers should use automatic matching with caution and that it may be best for novices to avoid it. In either case, only use automatic matching after you've done thorough keyword discovery with a tool like Lyris HQ.
If your keyword list is pretty robust, automatic matching can be useful as a supplementary keyword-discovery tool, primarily for discovering new negative keywords. Turn automatic matching on to find out which off-target keywords Google considers relevant. You can then tell Google not to serve up your ads when these terms appear in search queries, which will improve the ROI of your broad-match and phrase-match keywords.
If you haven't engaged in thorough keyword discovery, don't gamble on automatic matching. Google needs a well-thought-out keyword list to even begin to figure out which other terms to bid on; without strong keywords, Google's algorithm is guessing in the dark.
What's more, automatic matching gives Google permission to spend up to the limit of your budget. If there's a big gap between what you regularly spend and your cap, Google could crank out a lot of automatic matching that ultimately eats up a lot of your budget and, to be fair, generates a lot more traffic than you are used to seeing. The only problem is, the traffic generated is not likely to be high-quality traffic that really wants your products and actually converts.
Bottom Line? Proceed With Caution
While we don't dismiss automatic matching outright, we do view it as a tool for experienced search marketers who are already proficient at keyword discovery. Novices should leave the feature turned off.
###
About the Author
Dane Christensen is the SEM manager for Lyris. He is responsible for optimizing the company's PPC advertising spending across seven different search engines.
Like this article? Subscribe to our monthly email newsletter for online-marketing best practices.
 |
Subscribe to this comment's feed
