One of the struggles marketers deal with when measuring B2B lead generation is isolating and pinpointing the true cause of revenue performance.
At the top-of-the-funnel people interact with content and these people get rolled up into accounts. Both people and accounts have separate metrics such as clicks (leads) and velocity (accounts), making lead generation measurement difficult in B2B.
Compounding this problem is the fact that often revenue performance looks similar across lead generation campaigns, making it difficult to decide which campaigns to scale or optimize.
In this post we’ll provide a brief overview of B2B lead generation measurement that goes beyond the traditional activity report. We’ll provide an approach to measuring B2B lead generation in terms of deeper funnel metrics like revenue.
The end goal is to confidently decide how to scale your lead generation efforts. Let’s begin with first step in how to do that.
So let’s get started.
Step 1: Get the Revenue Numbers With Attribution
It’s all about working backwards and following the money trail when it comes to lead generation measurement.
We’ve written about tying marketing to revenue extensively, so we won’t spend too much time in this area.
The thing to keep in mind is that a situation can arise when revenue attribution alone is just a starting point for understanding what to do next for optimization. So step one is getting revenue numbers by choosing the best attribution model get a high level view of marketing performance.
Step 2: Be Aware Of Your Unit Of Measure
Be aware of your unit of measure, i.e. the thing getting revenue credit. Are you attributing revenue to accounts, marketing campaigns, blog articles, ebooks, webinars, paid media ads, or SEO keywords?
The unit of measure is sometimes called unit of analysis, it’s simply the rows in your dataset. Often times in revenue measurement the unit of measure is accounts, i.e. this account generated X revenue.
It’s important to be aware of this because often times you may need to group your accounts or Form URLs (or whatever your unit of measure is) into segments in order to isolate the effect that accounts for revenue performance. We’ll touch more on this in Step 6.
Step 3: Choose Your Response Variables For B2B Lead Generation Measurement
Let’s say our unit of measure is marketing campaigns. The response variable is what you want to understand or improve. Usually it will be revenue generated within a time frame, for example, why did Q4 generate X dollars in ARR?
But it can also be conversion rates, pipeline generated, or MQL’s generated.
Here it’s important to choose a relevant time frame if you’re measuring countable units like dollars. Time frames are important as you may account for seasonality or want to compare different tactics deployed at different times which may impact revenue.
If you’re measuring conversion rates then choose the funnel stage you’re most interested in. For example, conversion rates from content download to product demos.
Step 4: Address Similarities In Response Variable
In this step we’ll use an example to illustrate.
Let’s say revenue generated is your chosen response variable and ebooks (form URLs) is your unit of measure.
Looking at revenue performance isn’t very helpful if it doesn’t vary much, such as the situation where there are several ebooks with highly similar revenue attribution. If you’re trying to understand which topics to produce ebooks for or which ebooks to scale with paid promotion, then revenue attribution isn’t the best response variable.
But you want to optimize for revenue so how do you do this?
When revenue numbers look similar it’s difficult to decide strategy other than to continue business as usual. This can also be the case for keywords, sponsored events and campaigns.
Addressing this requires searching for whether revenue (response variable) is responding to different effects such as age.
A simple example is two ebooks with similar revenue performance have different ages, meaning one is newer than the other. Both ebooks have generated $10K in MRR in the last year. But one of them is only 3 months old while the other one is 11 months old. This could be a sign that investing in the newer one could boost revenue quickly because it is clearly associated with a faster deal velocity due to how quickly it is able to generate revenue.
This simplified example shows that despite similarities in the value of the response variable (revenue) there are explanations that can help marketers decide on next steps and strategy.
Step 5: List The Possible Effects On Revenue (Explainer Variables)
In the previous example, age (days since publish date) was a possible explainer of revenue performance. This explainer variable is one of many potential reasons revenue performance varies.
Explainer variables are an important part of deciding on marketing strategy and campaign optimization. So careful consideration is needed to isolate the effect to strategize and optimize with a high level of confidence.
With the above example, age helps isolate an effect on revenue and it’s a great variable to include in your analysis. But it’s not age that is explaining the revenue performance, it’s something else such as content topic, dollars spent promoting it, or marketing campaigns the ebook was included in.
Some of these explanations are measureable in your CRM or attribution solution, while others may not be. Either way, marketers should always consider all the explanations and understand which explanations can be measured with data and which cannot be.
This helps marketers isolate and measure the effect, then begin testing.
In the next step we’ll look closer at the explainer variables that are measurable.
Step 6: Isolate Your Explainer Variable
After defining a list of measurable explainer variables it’s time to explore which ones are good candidates for analysis. The point of this step is to isolate the 1-2 explainer variables that will be most promising to you for explaining the variation in revenue.
Look for group differences by grouping your accounts or ebooks (unit of measure) by age, marketing campaign, keyword, dollars spent promoting, and etc. If one group of accounts or ebooks is disproportionately a part of one campaign, receives more in promotion, or is in a certain age bracket then this could be an effect that explains revenue.
We’re looking for group differences because revenue is the same for this group. We’re looking for groupings that are unique to one ebook or segment of accounts as potential explainers of revenue.
Let’s look at an example where accounts is the unit of analysis.
There are two segments of accounts that have generated the comparable amounts of revenue. One segment of accounts is made up of fewer accounts but each account is a large deal size from a specific region of the country. This signals to you that whatever marketing activity you were doing to target this segment may be responsible for generating revenue.
Next, if possible, examine covariance and correlation between this variable and revenue across all accounts or ebooks (or whatever your unit of measure is).
Covariance is a measure of how much one variables changes when the other variable changes as well. For instance if an ebook gets $100 additional dollars in promotional spend then revenue attribution increases by $1000. Covariance is how two variables change with one another.
Correlation is the strength of the relationship between your variables and is measured in a standardized number between -1 and 1. Due to the use of standardized units it makes comparison easy if you run multiple correlations. A correlation statistic close to 1 denotes a strong positive relationship.
Sometimes there isn’t enough data or special attention needs to be paid to the type of correlation measure to use.
For example, inclusion in a successful account-based-marketing (ABM) campaign can explain why some accounts get more revenue than others but campaigns are a categorical variable and a specific type of correlation measure is needed in this situation.
To keep it simple, Step 6 in B2B lead generation measurement is isolating a few explainer variables and getting a measure of correlation and covariance.
Step 7: Decide Whether There Is Room To Scale
Continuing our example of a group of accounts or ebooks that are getting similar revenue attribution, it’s time to decide which campaigns and content to scale.
In the easiest scenario you find that ebook A is newer and was included in a new ABM campaign and that is the main difference for explaining revenue performance compared to the ebook B. This could signal you to:
Include ebook B in an ABM campaign can be result in more revenue
Target audiences may prefer ebook A because it is associated with faster revenue velocity
Invest more money in the ABM campaign promoting ebook A because the ceiling is unknown
This is a simplified example but the steps explained above will help you make these types of decisions with greater confidence.
The main idea here to search for differences between campaigns or ebooks that produce the same revenue performance, and to isolate the explainer variable.
There are certainly more steps that can be taken using more advanced statistical analysis but hopefully this is the start you need.
Measuring top-of-funnel lead generation efforts in B2B isn’t easy with the long sales cycle. It creates situations where revenue numbers look similar for many top-of-funnel units of measures like campaigns and content. It can also create situations where the best path forward is unclear.
But hopefully the steps outlined in this post provide a starting point for approaching measurement and optimization.
Start with revenue attribution and then work your way backwards to isolate the effect that influences revenue.
Why start with revenue and attribution?
Because success of optimization and scaling efforts will be based on the business-relevant metric that is revenue.
Connecting marketing to revenue is what we call Pipeline Marketing. Read the rest of the guide below for a complete overview.