Top 10 PPC Campaign Mistakes

Don’t let common, easily rectified mistakes derail your PPC (define) search advertising campaign. Your campaign may be suffering from neglect as you go off and chase the sexy, hot, shiny new forms of online marketing such as Twitter or video advertising. Whether your campaign has been up for years or is still fairly new, I’d bet there are still significant improvements you could make – plus mistakes you should correct – to compete in the “quality score ecosystem” that evolves monthly and is quite different from that of even a year ago.

One really fun thing about having been in search engine marketing for 16 years (yikes) is that by now, I’ve seen dozens, perhaps hundreds, of mistakes made in paid search campaigns. I’ve even made more than my fair share of mistakes while getting my hands dirty in the campaigns I run. One common theme I’ve seen in the campaigns of hundreds of marketers is that many foundational elements of the campaign are lacking, resulting in large missed opportunities. Failure to get the basics and foundational elements of a campaign right will reduce your Quality Score and also prevent you from engaging in the level of tuning required to make a campaign really hum.

All search marketers should go through this checklist of common campaign mistakes to make sure they aren’t wasting money or missing an opportunity. While the relative priority of the following list may vary slightly by advertiser, these cover most of the big mistakes that end up in old campaigns and even in new campaigns that have just gone live.

1. Failure to track, track the right things, or track at the right level: You’d think I wouldn’t even have to mention tracking and metrics, but sometimes marketers don’t measure and track at all, or they don’t track the things that really matter. The closer you can get to tracking immediate and long-term profit, the better. Sure, the return on ad spend (ROAS) metric is a good proxy if your margin is consistent and none of your customers have different life time value (LTV) profiles. But more often than not, there’s a good level of variance in margin and LTV.
2. Leaving on the Google Content Network without testing: Contextual advertising can be valuable, but it often works best in segregated campaigns. Just because the default is to leave content on doesn’t mean it’s the right decision for you. I recommend you first tune search and then go back and figure out how to make content work for you.
3. Foundational keyword research and expansion: Keyword research never stops. While there’s a point of diminishing marginal returns on adding additional exact match keywords to a campaign, most businesses change over time, a fact requiring a review of keyword opportunities. Seasonal businesses or ones where products are constantly being added to the site require even more vigilance.
4. Right mix of match types, including negative match: Exact match is great for getting Quality Scores high and therefore being able to afford high positions less expensively. However, you still need some phrase and broad match listings to catch the searches that the exact match listings will miss. Often one can learn from the broad match listings the keywords that deserve their own ad groups.
5. Proper account structure and ad groups with tuned creative: It’s likely that the Google schema of campaign to ad group will become the industry standard due to Google’s leadership position in the marketplace. After all, no advertiser wants to rebuild a campaign structure from scratch for each engine (regardless of whether Yahoo’s Panama disappears in a year or so or not). So revisit your campaign to make sure you haven’t taken shortcuts such as cramming too many keywords into an ad group. This problem becomes exacerbated when the keyword phrases look similar in meaning. Often the ad copy no longer includes the optimal phrases that were searched on. Search scent is strong, and searchers look for ads that are an obvious match.
6. Bids managed at the keyword level: Each keyword phrase is in its own auction, and bidding decisions should therefore be made granularly (except in the cases of extremely low-volume keywords). Just because you can set default bids at the ad group level doesn’t mean you should.
7. Landing page testing and pages selected at the keyword level: If user intent is different between two keywords, make sure the keywords get different landing pages – particularly if you have them available on your site already. The landing page that is best may not be the one that has the exact thing the searcher was looking for. Sometimes they want relevant pages with related options listed.
8. Failure to test ad creative within important or high opportunity ad groups: You don’t have to test every ad. But the larger the current and potential volume, the more important it is to test ad creative. Just a small improvement in CTR (define) will have a dual benefit of more clicks and better Quality Score.
9. Forgetting to test the marginal value of bidding for your own name and brands: People continue to ask me if they should bid on their name when they already have top organic position. Do a pulse test. Generally the answer is yes: you should still bid on your brand terms even with high organic position because you can more precisely control the user experience and also improve both total click volume and conversion rates on the paid clicks.
10. Using the engine’s budgeting tools: Only you know the true value of a click. Don’t let the search engines send you less valuable clicks (with lower profit) due to a budget setting. The budget settings are there for emergency purposes.

Bonus mistake: failure to test geotargeting and dayparting. It’s not that hard to figure out. Should you bid more by time of day and geography?

Got all your campaign fundamentals done and optimized? Then you can consider the cool stuff like search retargeting, video ads, and display ad enhancements.

This column originally appeared in the March 2010 edition of SES Magazine.