Once enough data has accumulated, it's important to run reports and take action based on the data. But what do all those numbers mean, and how should we make adjustments to improve PPC campaign performance?
Time to roll up our sleeves and dive into some data. Once enough data has accumulated, it’s important to run reports and take action based on the data. But what do all those numbers mean, and how should we make adjustments to improve PPC campaign performance?
For this exercise, let’s assume we’ve run a keyword report for a 120-keyword ad group from the campaign of leading widget manufacturer Widgets “R” Us. We’re going to focus on these four keywords: green widgets, industrial widgets, cheap widgets, and widgets.
Here’s the ad associated with the ad group containing the keywords:
Industrial Widgets
See Our Huge Selection.
Fast, FREE Shipping!
WidgetsRUS.com/Industrial+Widgets
Assume that the landing page for this ad contains multiple design elements — graphics and words — and the main message is that Widgets “R” Us features lower prices than its competitors. Assume also that the target maximum cost per conversion is $0.76, and that the campaign needs to be managed tightly to that number.
First let’s reflect back on the most important data points for optimizing PPC campaign results: conversion rate and cost per conversion. For an advertising campaign with a motive of profitable sales, conversion data is key to judging whether the campaign is reaching profitability targets. It’s also the key to making changes to the campaign and your site that can make it more effective at generating profitable sales.
But CTR is also an important metric. CTR is the gauge of how well your keywords and ads work together to get the click.
Conversion rate is the gauge of how well your keywords, ads, and Web site (especially the landing page) work together to get the conversion.
Broad View
CTR and conversion rates vary widely — the CTR range is between .02 and 2.92 percent, and the conversion rate range is between 7.41 and 38.81 percent. This tells us that the keywords in the ad group aren’t as closely related to each other, and to the ad copy, as they should be. The ad group is not “tight” enough.
The 120 keywords in the ad group should be split into smaller ad groups, with the keywords and ad text tightly related to each other.
Examining the Keywords
The keyword “green widgets” is performing well. There’s a high search volume, producing a significant number of conversions at exactly the cost per conversion that Widgets “R” Us is targeting. The 1.61 percent CTR is relatively low, though.
If that keyword (and other keywords in the 120-keyword list) were split off into a separate ad group that included the words “green widgets,” the CTR would improve. Then the advertiser could probably afford higher bid prices, which would improve the ad position on the search results page. If the advertiser also created a landing page with a headline and descriptive text using the words “green widgets,” the conversion rate would likely improve, the cost per conversion would decrease, and the bid price could be increased — hopefully allowing Widgets “R” Us to get a bigger number of profitable sales.
Let’s move to the term “industrial widgets.” That keyword is featured in the ad’s headline, so it’s no surprise that it’s earned the ad group’s highest CTR: 2.92 percent. On the other hand, conversion rate is lower than some of the other keywords.
The keyword would show much stronger conversion performance if the advertiser included it in an ad group with other related terms, and pointed the ad to a landing page with contents reflecting the meaning of the terms. Even just a headline that included “Tough Industrial Widgets” would spur better conversion rates than the landing page that features competitive prices. It’s likely that cost per conversion would drop, too, coming more in line with the target $.76.
The next keyword, “cheap widgets,” has garnered a CTR that’s just OK (1.86 percent) — reflecting the fact that the keyword doesn’t appear in the ad text. It exhibits the best conversion performance, though — because the landing page is all about competitive price. Split off into a separate price-themed ad group, it’s likely that the keyword’s performance would be stellar. The high conversion rate and low cost per conversion would warrant a nice increase in bid price that would move the ad higher on the search results page.
The last keyword, “widgets,” is a problem. Though the impression volume is high, CTR is an abysmal .02 percent. Conversion rate is low, too, which results in a cost per conversion that is much higher than the target.
The keyword shouldn’t be deleted from the ad group because any keyword that produces conversion is good. It’s possible that the broad match version of “widgets,” if moved to another ad group and paired with a more suitable ad, would result in a better CTR and conversions at an acceptable price. It’s also possible that the keyword is being matched with undesirable search terms (like “widgets in china”), and the advertiser could use negative keywords to stop the ad from being displayed when search terms unrelated to the site’s business are used.
In any event, the keyword’s bid price should definitely be decreased. This would likely drop the ad position such that the ad would only appear on the second page of search results — but that’s OK as long as the keyword garners some conversions at or below the target cost per conversion.
Listen to David sharing his PPC tips in a live Search Engine Watch Webcast, “Profitable PPC: The Fundamental Secrets,” on October 22, 2008.