No fluff - just the best news in paid search marketing every week.
Paid Search – Tips On Closing The Loop
Although paid search is far more trackable than many media – even online display advertising – many paid search campaign managers have difficulty tying detailed paid search data to actual sales. This is particularly true with long-sales-cycle campaigns, high touch lead generation campaigns, and campaigns that are designed to drive phone sales. Often, such campaigns track the cost of a lead or another metric that doesn’t necessarily tie them back to actual sales figures to get a sense of true conversions. This poses a problem in that uneven lead quality can lead to poor decisions in bidding, keyword selection, and so on.
In this article, I will discuss several ways companies can effectively attribute sales (or close the loop) in lead generation and phone sales campaigns.
Before getting started on this, you’ll need to:
- Make sure you are equipped to track. This may sound silly but many companies are still not tracking their online initiatives. With easy-to-use tools like Google Analytics, there’s really no excuse for not tracking paid search campaigns (or other online marketing efforts). Some website developers might also create havoc with redirects or content management systems that don’t fully pass identifying information from an analytics program (for example, the long gclid string from Google Analytics that contains, in encoded form, a wealth of data about the click that will be used to drive data collection about that user’s session and, hopefully, subsequent sessions). In such cases, there’s no substitute for nagging and repeating yourself
- Track appropriate metrics. To do this, identify solid goals and link them to appropriate metrics. You’ll usually identify a primary goal that’s tied to revenue in some fashion. If your company is attempting to get people to fill out an application form, a suitable metric would be to track completed application forms. Other metrics would not provide as clear a picture than if the metric is tightly associated to campaign goals. Secondary campaign goals are OK (don’t have too many of these), but make sure you understand the significance of such goals. In the example above, a secondary goal could be to track people who reach the second stage of an application process, while not necessarily completing it. In some companies, it might be a useful indicator of lukewarm interest from a prospect. In other situations, the connection between such interest and actual revenues would be low so the secondary goal should be given less credence in the campaign adjustment process.
Lead generation campaigns: lead quality is paramount!
All too often, these campaigns track cost per lead (CPL) and don’t tie information back to sales. In such a campaign, the information below is some data you may see:
Ad group 1
- Cost per month: 10K
- Leads per month: 35
- Cost per lead: $285.71
Ad group 2
- Cost per month: 10K
- Leads per month: 350
- Cost per lead: $28.57
Which one is performing better? Based on the above information, based on CPL, one may assume ad group 2 is performing better than ad group 1. One may even axe ad group 1 in favor or ad group 2. But the picture is actually incomplete. If we had taken a look at actual sales data, we would have learned the following:
Ad group 1
- Conversions (purchases) = 25
- Revenue from conversions = $100,000
Ad group 2
- Conversions (purchases) = 3
- Revenue from conversions = $1,000
So which is performing better now? Ad group 1 had a higher CPL but also a higher number of conversions and revenue attributed to conversions. If you had selected ad group 2, your ad group would have made $1000 as opposed to $100,000 for the same amount of money spent (10K).
Sound like an extreme example? Well, depending on the searcher’s intent, as represented by different keywords, and the business model, it’s not uncommon. Take a business model with a very narrow buying population, people who must be interested in the “pro” version of a particular kind of expensive software. The leads from one set of keywords are converting beautifully to sales, whereas the leads from another set of keywords are converting at nearly 0% because the campaign doesn’t make it clear enough that the software is not inexpensive or free, and because the keywords in this second group are too generic, targeted to a very broad audience. It may be possible to proceed with the second group, not throwing out the baby with the bathwater, by using some negative keywords and changing ad copy. But the second group is inherently worse than the first for this particular target audience.
Integration to incorporate revenue data into reports We couldn’t have had the above conversation without those revenue figures, then. Without them, we would have been flying blind. Here are several options to tie lead information to actual sales data by pulling paid search information into CRM systems. Using various CRM systems, Omniture, HitBox (HBX), Clicktracks, Webtrends and Google all offer options to link lead and sales information. In the case of Google, they allow integration with Salesforce to provide information on opportunities generated from Google AdWords like the number of prospects and closed/won business.
Here’s a procedure I recommend to tie phone sales to actual sales figures (inexpensive solution):
- After selecting a solution vendor, do some brief customization work and install code on your web site
- Customer calls to make a phone purchase
- Phone rep requests page ID from customer and enters info into system with ID
- Periodically, information is sent over to the manager of the paid search campaign (such as members of our team here at Page Zero). In long sales cycle campaigns, without any major reason to make daily adjustments, every 1-2 weeks is fine for frequency
- The campaign manager goes into the back end and inputs the phone sales data into a conversion tracking system
- Presto! The loop is closed
Good luck closing your paid search marketing loops!
Some opinions expressed in this article may be those of a guest author and not necessarily Search Engine Land. Staff authors are listed here.