When Bounce Rates Provide Best Performance Metric
If you’re choosing a professional PPC Manager, you’ve probably been bombarded with the term “ROAS” and “ROI.” Everybody wants a “Return on Advertising Spending.” Some PPC Management experts are even so bold as to characterize PPC ad spending as an investment. Those will be the ones favoring the term “Return on Investment.” Both terms refer to a critical form of analysis required for the effective management of a PPC ad account.
Stacks of Statistics
One of the great things about search marketing is everything which happens online can be tracked. All the tracking generates an impressive pile of numbers. From this pile, one can derive solutions for improving, even optimizing, ROAS. It’s involves measuring the cost of a conversion.
What is a conversion? A bid is placed on a keyword phrase, a searcher types in the phrase, your ad is served, the searcher clicks on your ad, lands on your website, and then performs some act, like complete a contact me form, or buy a product online. In this way a new customer or prospect has found you. That’s a conversion. If we can place a dollar value on the customer, then we are able to value the conversion event, and then calculate the return you earned on your PPC ad spending. Indeed, for many online stores, this conversion process is watertight and the return earned in connection with a new sale is calculated with certainty.
Not Everything Happens Online
However, there exists a not-so-widely discussed little fly in the ROAS ointment: for most PPC advertisers, not everything happens online. The most prevalent example is conversions by phone. If the prospect mentioned above phoned you instead of completing your online form or purchasing something online, then you suddenly have a relevant conversion which is not captured online by your pile of numbers.
Of course there are online solutions for capturing conversions by phone, and we have implemented such solutions. However, unless you are spending $1,000 per day, such solutions may not be economical for you. In fact, for most of our PPC customers, phone conversions represent a significant percentage of total conversions. For them, calculating ROAS requires an account manager to formulate some estimates and assumptions about conversion rates.
PPC Yin Yang 
Therefore, in calculating ROAS or its equivalent, for customers with phone conversions, we apply a two-pronged approach. Most of our PPC customers have websites which are designed to convert. The phone number is prominently displayed on every page, and a contact form is at hand or just a click away from every page. For such customers, we are often able to extrapolate the relative total conversion value of web pages and ad campaigns from the online conversion values. We do this by applying an estimate of the ratio of phone conversions to online conversions. This varies by industry and client. A widely-used ratio is 7:1. Absent any concrete evidence, we tend to apply a more conservative ratio of 3 calls to 1 completed form.
Of course, we strive to improve our estimate for each client over time, based on any monitoring of call rates which may be performed by our clients. So, extrapolating phone conversions from online conversions represents the first prong in our approach to evaluating ROAS performance for advertisers with unmeasured forms of conversions.
Bounce Rate Balance
The second prong relates to bounce rates. Often bounce rates are inversely related to conversion rates. Indeed, a high bounce rate generally precludes an advertiser from earning a high conversion rate. Over time, we develop estimates for each client about the relationship of bounce rates and conversion rates. In fact, for each client, one of the statistics monitored and included in our “PPC Ad Manager Roadmap” is a forecast for bounce rates.
When we create some new ad group, ads, or campaign for a client, we monitor the bounce rates and compare those to other campaigns or ad groups to evaluate the relative performance of the new ads. As part of this analysis, we calculate the “cost per net click” (CpNetC) of visits, by subtracting bounces from visits. This is an imprecise substitute for calculating ROAS based on conversions, but our experience shows that for many clients, bounce rates provide the best quick statistic for measuring the performance of any new advertising initiative.
