5 methodologies to measure the ROI of marketing
Before selecting methodologies, you’ll need access to the analytics for anywhere your marketing team spends money. From there, it’s a matter of using the right data.
While selecting methodologies require work upfront, your work can pay off in spades. According to Forrester Analyst Tina Moffett, B2B companies see an average 15-18% increase in revenue after taking a more sophisticated approach to analyzing and optimizing their marketing programs. That said, here are five methodologies you can use to track your ROI.
1. Baseline method
The baseline method is a simple calculation to determine your ROI on specific channels. First, calculate the cost per lead (CPL) on a channel. Take the amount invested on a channel and divide it by the number of leads produced by that channel in your defined time period.
Then you’ll need to get your baseline number. Multiply your lead to sale conversion rate by the average customer value (can be ARR, ICV, or LTV). If your CPL is less than or equal to your baseline, you have a positive ROI. Baseline should be either more than or equal to CPL.
As you can guess, the higher the difference between your CPL and Baseline, the better. With this method, you can compare the ROI of different marketing channels. For example, you may find it costs you $500 to acquire a customer through ads and $1,000 through events.
2. Experimental versus control groups
With experimental versus control groups, you compare the control and experimental groups against each other in an experiment.
To do so, split a group of prospects in half. Send one half a variant (the experimental group) to a marketing activity; leave the other half (the control group) using the current system as a constant. Identify a variable(s) to measure over time whether it's top of line metrics or MQLs. Plot each group on a graph and track over time.
By completing these experiments, you can measure how effective different marketing activities are over time and see the impact changes have on your KPIs by testing your experiment on a smaller group. Let’s say you want to see the effects of different account-based marketing activities. You may test adding personalization based on the industry of your target persona while keeping the rest of the content the same. So you decide on A/B testing, adding their industry to the headline on a landing page to see the effect on scroll depth and conversions.
Other follow-up experiments you can run:
Update the testimonial to someone more relevant to your target
Adjust the copy so it focuses on the core challenges of a job role
Display companies you have as customers that are more relevant to the industry of your target customer.
You can also run multivariate tests where you compare how effective a drip email campaign is versus another.
3. First touch attribution
An attribution model establishes rules that state how much credit different touchpoints in your funnel receive for sales and conversions. Under this attribution model, the first click matters most and gets all the credit for a purchase.
First touch attribution can help you see what top of the funnel content delivers value. Here are a few scenarios where it makes sense to view your campaigns under this lens:
See which blog posts on your site land a tracking pixel on leads that end up converting
Discover the referral channel with the most traffic that converts
A visitor might bounce around from one marketing campaign to another
4. Last touch attribution
Last touch attribution, also known as last click attribution, gives the final touchpoint a lead interacts with before buying all the credit for a conversion. By default, Google Analytics's Acquisition reports attribute completing goals to the last interaction. This attribution model can help you see your most impactful bottom of the funnel marketing activities.
For example, let’s say your lead sees the following before buying (in this order):
- Blog post via Google search
- Retargeting ad on LinkedIn
- Facebook ads
Facebook receives 100% credit for the sale.
5. Multi-touch attribution
Multi-touch attribution models distribute credit evenly across each touchpoint. This model is powerful for longer sales cycles with many touches. Running this attribution model for 3-6 months, marketers can optimize every touchpoint and turn off keyword ad groups that aren’t driving sales.
Let’s say a lead is worth $1500 over the next year. First, they discover you on a LinkedIn ad, read a blog post, and then sign up for a webinar before they buy. Each interaction has a $500 value by default under this model. You may find that leads who first find you through LinkedIn ads end up only booking demos, but do not purchase. Then you can reallocate your budget.
“A lot of times when companies use last touch attribution, they saw their brand’s AdWords campaign was driving all their conversions. So they dump money into that,” said Dan McGaw, founder of McGaw.io in an interview on the Marketing Analytics Show. “Come to find out once you run multi-touch, you find that it is not as valuable as you might think. We see a lot of companies distribute that spend from brand key terms to more middle of the funnel where they can actually make more money.”
Ruben Ugarte, Data Strategist at Practico Analytics, has helped 70+ companies use data to make higher quality decisions. Here's what he's found:
Companies will start with last touch until they reach low millions in annual spending.
At that point, they can look into upgrading into a multi-touch or a logarithmic model.
Channels that are easier to track tend to get more resources. This usually leads to higher spending on Facebook and Google, but less investing in word of mouth even though the latter may be cheaper and more effective.
Some companies build more advanced multi-touch attribution models that assign value to different touchpoints based on: