SearchPaid SearchPPC, Backwards and Forwards

PPC, Backwards and Forwards

In paid search, past performance isn't a true indicator of future results. Three ways to improve your chances of winning.

Traditional Web analytics may give you the wrong signals, steering you to the wrong decisions about your PPC (define) search campaigns. Unfortunately, most marketers’ decisions about marketing PPC and other on- and offline marketing campaigns are based purely on historical data. That’s like driving in a race using an oversized rearview mirror. Sure, your mirror will tell you a little about the curves in the road and traffic’s general position, but trying to drive competitively at high speed while looking behind is a formula for disaster.

My team and I have found that building on campaign data and overlaying marketing objectives results in a forward-thinking plan with a good chance of hitting these future objectives instead of getting mired in a scenario where positive change seems nearly impossible.

Yet Web analytics companies, ad tracking systems, engine-based analytics, and campaign management systems place a huge emphasis on what happened in the past. Fancy, high-sizzle interfaces reveal tons of information about keyword-level detail regarding CTR (define), conversion rate, and possibly revenue, profit margin, and whether the customer is new or returning. To win the race, in addition to looking forward at your objective, you must know who’s next to you, especially in a heavy competition. In the competitive PPC environment, those next to you are bidders who in real time are sometimes ahead of you and sometimes behind. These bidders have a greater impact on your success than how your ad performed in the past.

No one can argue that a lot can be gleaned from prior data, particularly if data is analyzed on a macro level as well as on a keyword-by-keyword basis. Some bid management systems even draw analogies to how analytics from the stock market play a role in decision making. Unfortunately, like mutual finds and stocks, there’s a limit to how predictive a model can be using purely archival historical information.

Indeed, historical data is a campaign strategy’s foundation. But to truly win in an auction market, one must factor in real-time data and plan experiments that allow one to maneuver quickly as the marketplace changes. This requires testing and probing the markets for elasticity on high-volume keywords. This elasticity testing can be performed automated, manually, or by using some combination of both methods.

Don’t Fixate on the Bids

Take a deep breath and remember what the bid is designed to accomplish. Higher bids are means to ends that include greater visibility, more clicks, and more business. But ROI (define), profit, and other success metrics restrain us from bidding to higher positions within our bid management systems. We fight with other marketers whose bids may be higher or lower than ours (given relative quality scores), regardless of whether these competitors are bidding rationally or not. Our objective is to determine how to raise our reserve price or otherwise give us an edge. Instead of fixating on the bids, we need to set an objective of raising our reserve price through conversion improvements or revenue/profitability improvements.

Enhance Conversion Rates

Use the data you have to plan a roadmap showing exactly how much lift in conversion rate you’ll require to make a material change in position over time. For example, the results of your elasticity testing may indicate that a 32 percent increase in CPC (define) is required to jump two positions in the SERP (define). To keep ROI constant, you need to figure out how to raise conversion rates by 32 percent. Of course, the difference in lead or order volume produced by your listing being two positions higher in the SERP will be fairly dramatic, often double or more (particularly if you move from the right rail to the top). However, vigorously monitor conversion rates, because a jump to first position will sometimes reduce the rate. You might think that increasing your conversion rate by 32 percent would be challenging. Yet a conversion lift from 3 percent to 3.96 percent will get you there. If you can’t find this kind of lift anywhere in your campaign, look at clusters where conversion is already there by geo or daypart.

Enhance Revenue and Profit

Conversion rate isn’t your only leverage to a higher position. Profit or revenue enhancement works as well. Revenue enhancement can be driven by doing such simple things as changing your promotions. For instance, instead of offering free shipping at $50, make it free at $59 or $65. The average order size may jump, giving you immediate leverage in your bidding strategy. Once again, if you can’t raise your revenue across the entire clickstream on high-potential keywords, consider a segmentation analysis.

In 2008, let’s all resolve to plan and execute campaigns by looking forward, setting objectives, and determining milestones required to achieve these objectives. And don’t forget to take the SEMPO State of the Market Survey.

Want more search information? ClickZ SEM Archives contain all our search columns, organized by topic.


US Mobile Streaming Behavior

Whitepaper | Mobile US Mobile Streaming Behavior

Winning the Data Game: Digital Analytics Tactics for Media Groups

Whitepaper | Actionable Analysis Winning the Data Game: Digital Analytics Tactics for Media Groups

Giving a Voice to Your Brand

Whitepaper | AI & Automation Giving a Voice to Your Brand

Mobile Messaging Masters

Whitepaper | Mobile Mobile Messaging Masters