Measuring the success of a search marketing campaign starts with a fundamental metric: the click. But clicks alone don’t tell you the real story.

Where is the traffic coming from? How much did it cost to generate? What did those visitors do once they arrived at your site?

Similarly, measuring conversion rate – the percentage of search-generated visitors who make a purchase or answer a call to action, such as registering for a white paper – doesn’t give you enough insight to optimize your campaign performance. Conversions and clicks to ads or URLs must be analyzed in context. That’s why we asked experienced marketers to share their selections for underused or neglected metrics that can help refine a search marketing campaign.

Most of the five underused metrics were also top suggestions in our recent Search Marketing Benchmark Guide. We’ve included them with an explanation on how to calculate each metric, and added suggestions for putting the data to use.

The Five Metrics

-> Metric #1. Ad spend ratio

Description: Ad spend ratio calculates how much of each dollar in revenue you spent on the PPC ad that generated it. Calculating ad spend ratio daily or weekly for each keyword can show which costs more per click than it returns in profit.

“When we compare how much we spend on that word to how much sales came in on it, we can quickly see if we’re spending more than we’re getting in,” says Ray Harris, Chief Campaign Architect, SitePropeller, Inc.

How to calculate ad spend ratio:

- Use your PPC campaign management engine to run a report that shows you the amount of money you spent on each keyword.

- Calculate the revenue generated by those keywords. (Google Adwords can do this. Yahoo! and MSN systems can not, requiring markers to use another sales-tracking system, such as a pixel tracking program, to record sales from those engines.)

- Divide the cost figure by revenue figure for each keyword. The result is the percentage of revenue you spent on each keyword – or ad spend ratio.

- Compare the ad spend ratio for each keyword to your profit margin on the goods or services sold to determine if it was a profitable keyword. The lower your ad spend ratio, the better.

While every company’s cost of goods will vary, Harris offers these percentages as a rule of thumb:
o 10% ad spend ratio for hard goods that must be shipped
o 50-60% ad spend ratio for intellectual property or online services that don’t have manufacturing and shipping costs

Use ad spend ratio to:

#1. Manage your keyword bidding strategies by looking for high and low extremes in your results. If you find certain keywords with an extremely low ratio, examine your ad position for those terms. If you are below the top three ads, you have a chance to increase your bid and gain a better position without harming your profitability.

Conversely, if you find keywords with high ratios, you can either suspend bidding on that word or try lowering your bid to see if you can maintain traffic while lowering your ad spend ratio.

#2. Double-check decisions made on CPC metrics.

Harris recalls a client who had suspended bids on certain keywords because the cost-per-conversion for those terms was higher than average. However, by calculating ad spend ratio, he saw that those keywords generated much higher total revenue and, therefore, were worth paying a higher CPC.

-> Metric #2. Bounce rate

Description: Bounce rate measures the number of visitors who either leave your site immediately after arrival (within a few seconds) or don’t navigate to any pages beyond the one they arrive on. High bounce rates usually indicate there’s a problem with the quality or relevance of the information on that page, or that the navigation options or layout of the page are unclear.

“I can be driving new visitors to my website, but what are they doing when they get here? As a marketer, that’s really what I’m interested in,” says Brandon Doty, Director, Marketing, The List.

How to calculate bounce rate:

- Most Web analytics tools will offer bounce rate as an option in their metrics list.

- Run a site-wide bounce rate report to generate a baseline for comparison.

- Run focused bounce-rate reports based on specific criteria, such as for PPC landing pages, most popular pages on your site, or by different traffic sources.

- Use filtering techniques to avoid skewed results from frequent visitors who may only visit one page at a time, such as employees or clients. You can either:
o Block visitors according to IP address to strip out results from internal visits, partners or clients
o Filter results by new visitors vs. repeat visitors

Use bounce rate to:

#1. Determine whether ads are pointing to the right page.

A high bounce rate could indicate that the content of the landing page isn’t matching what the visitor expected to see based on the search term used.

Shane Daley, Manager, Search Engine Marketing, Decorative Product Source, uses bounce rate analysis to refine his ad strategies. For example, the team had been sending very specific search terms, such as “single bowl kitchen sink,” to their site’s general sink category page. After seeing high bounce rates, they began pointing those clicks to the sub-category of single-bowl kitchen sinks, generating a significantly lower bounce rate.

#2. Highlight landing pages that need optimization.

A high percentage of search visitors who leave your landing page immediately signal that the page is not engaging them in a meaningful way.

Begin landing page optimization tests. Look for improvements to layout, call-to-action, copy, and other features that will influence visitors to stay longer and click through.

#3. Improve content, layout or navigation of internal pages.

An internal site page with a higher-than-average bounce rate could include factors that prevent visitors from moving forward from there. Problems could include:
o Copy that’s not relevant to the search term that delivered the traffic
o Confusing or too few navigation options
o Page layout or design
o Metatags that don’t match the page’s content

-> Metric #3. Click path analysis

Description: Click path analysis monitors a search visitor’s browsing pattern that led to their arrival on your site, and then the sequence of pages visited as they navigate within it. Patterns in browsing behavior can reveal how your customers are searching for your product or services online – and what other websites they’re visiting in the process.

How to calculate Click path analysis:

- Some Web analytics and PPC bid management systems provide external click path analysis. It shows the sequence of sites a visitor navigated through before arriving at your website.

Look for trends in searching behavior, commonly visited external sites, and unusual referring sites, which may contain clues about new keywords or demographic profiles to target.

- Use your internal Web analytics system to monitor where visitors click and what pages they visit once before they convert or exit.

Use click path analysis to:

#1. Gain insight into what influences a search visitor to convert Stephanie Chiquette, Marketing Manager, Albert at Bay Suite Hotel and Best Western Victoria Park Suites, confirmed through click path analysis that visitors looking for lodging in the Ottawa area often search a range of different terms. They may visit her websites multiple times while also searching third-party information before ultimately converting.

“The reports reconfirm that, yes, these are the sites they search,” says Chiquette. “They not only Google us, but they Google us again and often use different search terms.”


The analysis uncovered third-party sites to target for listings and helped Chiquette determine which keywords are influential in a preliminary search and which lead to a more direct conversion.

#2. Look for navigation improvements that can get to a conversion page in fewer clicks.

Identifying trends in the information visitors seek before converting can help you add content to key pages, or move them to a conversion more quickly by developing new linking strategies for the typical browsing path.

-> Metric #4. Lifetime value of a search visitor/customer

Description: Lifetime value is a tally of the total revenue generated by a search-referred customer, including immediate sales, delayed purchases or estimated future purchases/contract renewals. Looking beyond direct revenue to calculate lifetime value can create a more complete picture of search ROI.

How to calculate lifetime value:

- Create estimates based on your past business experience using formulas, such as multiplying the average value of a sale by the estimated number of times a customer buys. B-to-B service companies or subscription marketers may estimate the average value of a contract and multiply it by the average length of the relationship.

You can use those estimates to plan search campaign budgeting or set guidelines for acceptable cost-per-click expenditures.

- Track delayed sales or analyze the different LTV of search visitors by source, such as referring search engine or keyword used. Tag those visitors with a cookie when they first reach your site and retain data on future visits and conversions.

B-to-B marketers can use labels in their CRM systems to indicate that a new prospect first entered their database via search. Then track those prospects’ progress through the marketing and sales cycle to attribute subsequent revenue to a search-originated lead.

Use lifetime value to:

#1. Expand search marketing strategies to include new product categories.

Many marketers, particularly in B-to-B firms, avoid using search campaigns for high-priced items that don’t convert immediately. Knowing the lifetime value of a search-generated referral allows you to justify search expenditures as a way to bring prospects into a conversion funnel.

John Thyfault, Principal, Thyfault & Associates, created a search campaign for a client selling laboratory equipment that cost tens of thousands of dollars. Because of the high product cost, there was very little competition for those keywords. The team was able to create a campaign that used low-cost search ads to generate catalog requests.

By coding those catalogs according to the search-generated request, they were able to attribute sales generated several months later back to the search campaign.

#2. Refine keyword strategy.

LTV analysis can uncover certain keywords or ad groups that may not deliver an immediate conversion. But they may have high long-term sales potential and justify inclusion in a campaign.

#3. Determine the point at which keywords become unprofitable.

Jon Eggleton, VP Marketing, AG Interactive, faces increased costs from rising competition for popular keywords. At some point, he must decide when to abandon a keyword because of poor ROI.

“We can continue to spend money to acquire that customer, but campaigns are going to become less profitable,” says Eggleton. “Lifetime value helps us decide where that line should be.”

-> Metric #5. Profit per keyword

Description: Profit per keyword ties clicks and conversions for each word to back-end financial data, such as revenue generated by those clicks, to calculate the return on your search investment. It can be used in conjunction with lifetime value estimates.

How to calculate profit per keyword:

- For each keyword in your campaign, multiply your total clicks by your cost per click to calculate your total expenditure.

- Subtract that total expenditure from the revenue generated from search clicks.

- Divide your total number of sales or conversions into that total to calculate cost per acquisition.

Use profit per keyword to:

#1. Find your most profitable keywords.

Eggleton of AG Interactive uses keyword profitability metrics to decide where to spend his search marketing budget. If he notices a keyword delivering customers at a relatively low cost per acquisition, he’ll move more money into those terms to acquire more customers at that cost.

Conversely, he’ll move away from terms with strong conversion rates if his calculations show that he’s spending too much to bring those visitors to his landing page.

“If you look at conversion data, it can be very misleading,” he says. “Cost per click and other metrics are somewhat meaningless if you don’t tie them back to what you’re selling from those clicks and the total value of those sales.”

Useful links related to this article

AG Interactive
http://www.aginteractive.com/

Albert at Bay Suite Hotel
http://www.albertatbay.com/

Decorative Product Source
http://www.dpsource.com/index.html

SitePropeller, Inc.
http://www.sitepropeller.com/

The List
http://www.thelistinc.com/index

Victoria Park Suites
http://www.victoriapark.com/