How to Analyze Your PPC Lead Data (With 15 Mini Case Studies)
If you have been properly optimizing your pay-per-click (PPC) campaigns, or have an agency doing it for you , you should be generating leads. After all, that is why you are advertising…right? What most advertisers (and agencies) forget about is optimizing after the phone call or form submission.
What if lead quality is poor?
What if you can’t close a sale?
Not all leads are created equal, and there are several ways to reduce low quality leads. However, you must learn from your leads.
Once you start tracking your leads through the PPC lifecycle, you will need to gather enough data to be able to segment. You may now ask – how much is enough data? This really depends on how you segment your data.
Below are 15 different ways to analyze your PPC lead data with some guidelines on how much data you need. While the technique of analyzing these segments may seem redundant and findings may be linked with other analysis, the point is to show a small portion of the many ways you should look – and learn – from your data.
Want to dive deeper? Learn more about analyzing your PPC lead data by downloading our 10-page guide that outlines how to bridge the data gap between leads and sales to gather insights and drive action.
Length of Sales Cycle
It’s extremely important to consider the length of your sales cycle when segmenting lead quality. That’s because a business that closes leads in one day will analyze data differently than a business that takes 6 months to close a lead.
For example, a solar company that typically closes a lead in a week will need to take that time into account before determining lead quality. If a potential customer clicked on their ad and filled out a “request for information” form, they’ll need to wait until their sales team has had a chance to talk to that person, set-up an appointment for an estimate, and even purchased or leased the solar equipment before they can gauge the true quality of that lead.
Phone vs. Form
One of the many questions we receive regarding landing pages is:
“Should I put a form on the landing page or just a phone number?”
This question is usually followed up with “It depends.”
It depends first on the type of services or products you are marketing. If you provide urgent services, such as a DUI lawyer or a plumber, a phone number should probably be more prevalent, but not all services are that easy. For the rest, we recommend both, then we can learn the intent of both phone calls and form submissions and therefore attribute value to those actions.
We analyzed 8 months of lead data from one of our clients, a solar company. The call-to-action was “request a free quote.” Because it would be reasonable to assume someone may submit a form or call for a quote, both were on the page.
What we found was that people who picked up a phone and called turned from a lead to a sale at a rate almost 29% higher than form submissions. From here, we focused more of our efforts to drive phone call leads to increase close rate and increase revenue – the real reason you are spending money on advertising.
Day of the Week & Time to Follow Up
Like you would segment your PPC data in AdWords or Bing by day of the week, this is a great way to know what day leads are more likely to purchase. For example, for one client we looked at, the close rate on leads depended on when they became a lead.
As you can tell from the table above, weekdays have a significantly higher close rates than weekends. Now there could be several reasons why this could be: intent could be different between days, sales team working hours, and average time to follow up.
After presenting this data to our client, we were told that leads that came in during the weekend were not usually followed up by a sales member until the following Monday, compared to typically a same day follow up. Leads that were followed up with on the same day (weekday leads) closed at twice the rate of leads that were followed up between 24 and 48 hours of becoming a lead (weekend leads).
When thinking about how business relates to location, most people think about demographics such as income or interests. While this may be a good indicator for most businesses, it is important to look even deeper. For example, let’s say you own a plumbing business and your lead data maps to the graph included below with three different zip codes that you service.
One of the first observations you may make is that the average revenue per service in zip code 1 is by far the highest. Questions you should ask:
- Is this because they love spending money on plumbing? Probably not.
- Does it mean the customers in that zip code make more money? Maybe.
Without doing the actual plumbing work, you probably wouldn’t be able to make sense of this. It is possible that the plumbing issues in zip code 1 are more severe due to larger tree roots in that area, causing broken pipes and more expensive work. You may also find out that the sinks and garbage disposals in several neighborhoods in zip code 2 were poorly designed, causing issues more often, but a cheaper service. You can then target your ads more specifically to generate leads in certain zip codes that provide the best PPC ROI.
One of the more unique and interesting ways we have analyzed data is according to how leads fill out landing page forms. Having access to landing page data can definitely be beneficial in many ways, but have you ever thought about qualifying leads based on asking one simple question on your landing page form?
Take for example our client in the insurance industry. We added a simple question on the landing page form process asking “How Can We Help?” This question was fairly uninvasive to our prospects, but allowed us to learn a lot. We were able to score the leads that were sent to our client’s offline sales team to assist in prioritizing their follow up.
The form options and results can be seen in the table above. By collecting data based on how users filled out the answer to this form field, we were able to confirm that those who selected “just browsing” options were information seekers and were higher up in the funnel, while those who selected an option related to needing coverage had an immediate need, and they eventually lead to more revenue for the client. Read more about this in our full PPC Case Study, “Attributing Offline Leads to AdWords Sales.”
Depending on the amount of data you have for each, you can choose to analyze your leads by keywords or ad groups. In a perfect world, we would know the value of each keyword and to know that value we need to track sales back to the keyword. Fortunately for the same insurance client from the case study above, we had enough keyword data on their most basic keyword. The difference between the two keywords is their match type, which makes a significant difference. The results from the study below show the same keyword, one phrase match and the other exact match.
This is huge! That is an 80% difference in close rate between two keywords. Moving forward, the value for the exact match keyword is much higher than leads coming from the phrase match keyword.
While there are several ways you can A/B test landing pages, one of the most common is testing copy. One of our favorite tests is “qualifying” vs. “benefits” copy.
For example, let’s say you are advertising for a new service that a client has to meet certain requirements such as annual revenue, # of employees, etc. Qualifying copy may include something like “You may qualify for this new service if..” then a list of customer requirements. Benefits copy will typically just list the benefits a client would receive from the service.
This is an example where not analyzing your lead data could really cost you. This is because if you ignore close rate and only look at cost per lead from each page, you would likely move forward with the wrong page. The leads from the qualifying copy page close 2 times more often than the benefits copy page. Moving forward with the qualifying copy page could lead to almost 40% more revenue!
The difference in close rate is likely due to the qualifying copy setting better expectations. This is also the reason why cost per lead is higher: it weeds out more of the low quality leads.
Time of the day
Depending on your product or service, analyzing your lead data by time of the day may make sense for other reasons than how quickly you follow up with those leads. It may come down to the intent. For example, if you are a DUI lawyer, a phone call lead at 2am may be more likely to become a client than someone in the middle of the day. The lead coming in at the middle of the day is likely in the research phase and looking for a lawyer for their DUI last week. The intent here is different and may help you better spend your advertising efforts in the middle of the night.
Lifetime customer value
Knowing your lifetime customer value is important if you offer a product or service that customers are likely to use more than once or you have several products they can come back and purchase. To calculate this metric, you need to know the average value of a sale, the typical number of recurring transactions, and the length of time that someone usually remains a customer.
After looking at an entire years of data for a training video client we found that on average, a customer would purchase 4 videos throughout their lifetime. This can give you more leverage with your cost-per-lead goals since a customer that is likely to purchase from you again in the future, has an overall value that is actually greater than the value of their initial transaction.
For campaigns with heavy budgets in display ads, it is important to look at where your leads are coming from. Answer these questions:
- What site were they on when they see your ad?
- Do the demographics of the visitors of that site matter?
- Does the content on that site make a difference?
For one of our clients, the goal was to generate email sign-ups for an email lead nurturing process that sold our client’s training videos. With a large volume of data spanning over a year, we noticed that some display sites sent leads more cost effective than others. But we also were able to determine an average number of repeat sales from each display placement. Also, as a customer progresses through the training videos, they saw an increase in price per video.
Looking further into the display placements on ehow.com, we found that the content on those pages were much more relevant to our client’s business. This lead to not only a higher average number of repeat sales, but because the videos got more expensive, the customers with more repeat sales also had an increase in average revenue per sale.
Type of Lead
When it comes to types of leads, such as someone looking for a free quote vs. a free guide download, it is important to know the value of each lead. In the example below, the close rate for free quote requests is significantly higher (as it should be) than free guide downloads. This again is due to the intent behind the actions taken. Only looking at close rate, wouldn’t you just focus your advertising budget on generating free quote requests? Probably.
Taking it a step further is where the difference in cost-per-lead comes in. Because the cost-per-lead is so low for free guide downloads, in theory, it would make more sense to allocate your budget to these leads rather than free quotes. There may be several other variables to consider before completely changing your strategy such as length of sales cycle.
Sometimes the intent from your leads is all relatively the same. When you are still seeing poor performance, look at your sales teams. After several months of running PPC campaigns for one of our clients, a franchisee of a larger home contracting company, we observed that they couldn’t get their leads to close.
When looking at the sales teams between their franchise and another, we noticed their close rate was significantly lower than the other franchise. The appointments were coming in at roughly the same rate, but the sales were low. When speaking with corporate, we found that Sales Team A had not yet received corporate sales training. This happened to be a broken internal process, not an advertising issue.
If you’re the type of company that offers multiple services, you may find that one service is more profitable than others. For example, if you do both plumbing and HVAC work, it’s possible that you receive more requests for one service, such as emergency plumbing, while other services, such as replacing an air conditioner, result in a higher profit. It’s important to know which services provide the most overall value so you can allocate more of your marketing money toward those services.
Advertisers often use different types of ads for different purposes. For example, search ads are a great option if you’re just getting started with your advertising campaigns and have a limited budget, while display ads are great to increase your brand awareness. Typically the best approach is to test using both types of ads and determine which generates the most conversions at the best cost. This also may imply a different intent from your visitors and leads from each campaign type and therefore have a different close rate.
These are just a few examples of how you can successfully segment your PPC lead data to determine which types of leads are converting better and are more profitable. Knowing what to look for and how to analyze your data appropriately will ultimately help you make better business decisions that positively impact your bottom line.
About Keegan Brown
Keegan joins the team as an Digital Advertising Analyst with a diverse background in business. While receiving his Bachelor of Science in Accountancy, Keegan spent one year in a tech incubator creating a mobile app. His entrepreneurial spirit has led him to learn web development, sales and internet marketing. In his free time Keegan enjoys music, exercising and weekend trips to Flagstaff.
How Google Grants Can Help Your Nonprofit Succeed in AdWords
Jan 10, 2017
Google Analytics 101: Best Practices for eCommerce Sites
Dec 22, 2016
Marketing Automation Platforms and SEO: Subdomains vs. Subdirectories
Dec 19, 2016
Using the Canonical Tag and Redirects to Solve URL Canonicalization Issues
Dec 09, 2016