Mislead by Conversion Rates: A Closer Look at How Marketers Can Improve Campaign Performance
Recently, a leading multi-channel retailer engaged in a landing page test that sacrificed thousands of dollars and compromised plenty of ROI. The retailer’s fatal flaw? Relying solely on the almighty conversion rate.
The retailer tested two landing pages for its #1 keyword, “red jackets.” The first was a product list page featuring only red jackets; we’ll call this “The Red Page.” The second, “The Promo Page,” displayed an attractive young person modeling a jacket (not red), with a “Save $15 on orders over $100″ promotion.
The search engine marketing team alternated “The Red Page” and “The Promo Page” for all keywords in their Google “red jacket” campaign. Two weeks later, in an executive meeting, the search engine marketing manager boasted about the results.
“The Red Page proved to be the winner in our landing-page test with a 21% higher conversion rate than our Promo Page,” the manager said.
The announcement of the test results prompted smiles, “good jobs,” and other congratulatory verbal and non-verbal reactions by executives; none of whom fully realized they had just been duped by the misleading nature of the conversion rate.
Conversion rates are one of the most measured and optimized metrics in search marketing, and for good reason. A conversion rate is the percentage of visitors that result in a sale. So when a conversion rate improves, it is logical to assume that the ultimate metric, ROI, will also improve. Well, not so fast.
As a search engine marketing professional, I face situations like this all the time. Clients and other online marketers place too much emphasis on conversion rates when measuring performance. But conversion rates aren’t always king; king being defined as an “absolute ruler.” It is not absolutely guaranteed that one landing page, text ad, or other marketing component is performing better than another solely based on conversion rates.
While excited over the success of “The Red Page,” search engine marketer X neglected other factors that affect campaign success like traffic, revenue, average order value (AOV), earnings per click (EPC), and ROI.
Some visitors arrived on “The Red Page,” found a red jacket which retailed for $50, and went through the checkout process. Others landed on “The Promo Page,” found a red jacket for $50, and noticed the additional incentive, which offered $15 off of orders over $100.
The data below illustrates how other factors play a considerable role in search marketing performance. Had search engine marketer X considered more than conversion rates, the retailer would have maximized ROI and revenue.
Red Page: 2.3% Conversion Rate
Sales: 69
Revenue:$3,450
AOV: $50
EPC: $1.15
ROI: 2.3
Promo Page: 1.9% Conversion Rate
Sales: 57
Revenue: $4,845
AOV: $85
EPC: $1.62
ROI: 3.2
Lifetime value aside, the “Promo Page” is the true king in this test. If the retailer actually stopped testing the “Promo Page,” it would have forfeited $35 in AOV, $0.47 earnings per click, and a landing page that provided a 39% higher ROI. Not good at all.
Online marketers need to look at e-commerce conversion rates not as the sole performance indicator, but include it in a balanced scorecard that measures true online success.
Reader Comments.
Another way to improve your search marketing performance: Buy seemingly unrelated keywords.(http://johnrodkin.blogspot.com/2006/10/downside-of-quality-score.html)
Conversions can be quite misleading. Granular revenue data has to be taken into account especially when the promotion has a predictable effect on AOV like a spend X and receive Y.
Michael, thank you for raising a critically important subject for the industry as interactive marketing matures.
Clearly, we are interested in not just improving each hop in the conversion funnel. What we ultimately want to achieve is optimization across the entire customer decision making process.
At the moment, most companies struggle with this task given the functional silos that exist across lead generation, sales and management of the customer experience. This is exacerbated by further silos across channels.
The solution is integrated planning, testing and optimization across customer touchpoints. The benefits are obvious; profitable conversion of the right customers with deep insights gained about the decision making process itself.
Digital marketing is in fact leading us to a new functional paradigm where the customer facing process is the hub for a range of rapid test&learn cycles that inform key business processes.
All businesses need to transition from simple metrics that suit organization charts to more wholistic processes that center on the customer’s choices.
Coremetrics is a really good tool in figuring out conversions and drop off rate, intelligence is key to success for e-commerce sites.
Your article mentions “(2) lead quality percentage”. How do you measure this in practise ?
Thanks
Donald
The best case scenario is when you have the IT infrastructure capabilities to create a closed loop feedback system in which all leads are tracked throughout the entire sales process (click to final sale). This provides the greatest clarity, however is not always practical. Recently our agency introduced a new metric to our clients, CPQL or Cost-per-quality-lead. We actually develop a list of criteria with each client under which we agree a lead will be considered a “quality” lead. The criteria differ for each client, but some of the more common criteria include, (a) Valid Phone (b) Accepted Geography (c) Verified Need and (d) Allocated Budget.
Regardless of which system you utilize, there must be some connection to the online lead form and the lead once it enters the organization lead flow process. We have found one of the easiest ways is to create a unique ID for each lead generated at the time of form submission. This identifier can be passed as part of the online lead form to the sales team and easily passed back to your online tracking technology, which gives you that needed identifier to really begin appreciating the quality of your leads.
Hopes this helps, thanks for the great question..
Leave a Comment
Pages: 1 2 next page »
Tags: campaigns, lead_generation, ROI and SEMArticle Sponsor
More Features
-
Loading ...
Latest News
- Hulu’s Bringing Its “A” Game But… March 19th 2010 ADOTAS – Hulu’s sales team is actively subverting the ad [...] more »
- Yelp! A class-action suit? March 19th 2010 ADOTAS – One of three civil suits against Yelp filed [...] more »
- Viacom Accuses Google; Testing Digital Millennium Copyright Act March 19th 2010 Viacom has accused Google of turning a blind eye to [...] more »
- Google to Leave China April 10th? March 19th 2010 ADOTAS – According to the China Business News, Google Inc [...] more »
- [x+1] Creates The Smartest Tagging System Around March 18th 2010 ADOTAS – Today, if you happen to be at the [...] more »
- IAB’s Video Standards Tackled By ADTECH March 18th 2010 ADOTAS – ADTECH, a part of AOL Advertising and an [...] more »
- Google Search and Mobile and….TV? Oh My! March 18th 2010 ADOTAS – Google wants to dominate your screens…. Not just [...] more »
Features
- Growing Pains March 19th 2010
- For Better or For Worse? March 18th 2010
- Yahoo! Wants to Get More Personal March 17th 2010
- Creative Considerations for the iPad March 16th 2010
- Amazon Leaves Colorado Affiliates Out in the Cold March 12th 2010
Spotlight
AdBidCentral’s CEO, Vivek Veeraraghavan Talks Openly*What was the inspiration to start AdBidCentral? The conditions that inspired AdBidCentral came from a variety of factors in my personal [...] more...
Reader Favorites
Classifieds
- Communications Fellow
- Sr Director, Marketing Services
- Senior Web Analyst
- Account Director
- Director of Analytics
Recent Comments
- Jedd Gould: I think you miss the point. Publications have to charge because the content most are
- Gavin Dunaway: They're similar, but Ning is more a competitor to Facebook and MySpace while StumbleUpon considers
- Steve Feldman: Does StumbleUpon compete with and does essentially what Ning networks does?
- Bulent: I wonder how many of the clients would accept a media plan, that would -as
