Profile picture for user Jordan Con

With the right tools, the general concept of calculating the ROI of paid media is pretty simple.

A multi-touch attribution solution will tell you how much revenue is generated by paid media overall, as well as by specifics, like by campaigns, keywords, ads, etc. If your attribution solution has API integrations with the paid media channels, you can even pull in the cost data in the same place. If not, you can go to the paid media channels to get specific cost data and do the math in a spreadsheet. You can read more about calculating paid media ROI here.

In theory, it’s easy. But in practice, there are a lot of things that can get in the way of an accurate calculation -- especially if you're not using a centralized, multi-touch attribution solution. Here are 4 mistakes that paid media marketers make when calculating ROI, as well as what to look out for and how to fix them.

Mistake 1: Using a single-touch model

Single-touch attribution models are something that we harp on pretty often, here at Bizible. The reason is that they create a bias in the data called model bias.

We like to think of B2B marketing like a relay race. Every marketing effort, each touchpoint, gets prospects farther down the buying journey, with the finish line being a closed-won customer. Yes, some touchpoints are more important and move prospects farther faster, but getting to the finish line is a team effort.


relay race


When only one touchpoint receives 100% of the credit, that means all of the other touchpoints get left out. It’s like giving the gold medal to just one person on the winning relay team. It doesn’t make sense.

When it comes to paid media, your efforts may be targeted at different stages of the funnel. One campaign (Campaign A) could be targeted at people who have never heard of you before and another could be a retargeting campaign (Campaign B). Both campaigns may be contributing to moving prospects down the funnel, but if you use a first-touch model, Campaign A will always have an opportunity to get 100% of the revenue credit and Campaign B -- the retargeting campaign -- will never get any revenue credit, no matter how effective it actually is.

Without switching to a last-touch model, it would be impossible to evaluate the ROI of your retargeting campaign. And even if you did switch to a last-touch model, your retargeting campaign may still not show up because there could be plenty more touchpoints at the bottom of the funnel. See how this creates a problem for ROI calculations?

The solution is to use a multi-touch model that takes into account the full impact that marketing is having no matter where it is in the journey. For organizations that don’t do marketing past opportunity creation, the W-shaped model works great. If you want measure marketing past opportunity creation, a Full-path model is ideal. Both of these models measure marketing’s impact on the full funnel and give credit accordingly. This solves the model bias problem because it gives all marketing efforts a fair chance to get revenue credit.

Mistake 2: Not accounting for the full sales cycle

In the B2B buying process, it can take many months for a prospect to go from first touch to closed-won. If you try to calculate your ROI too soon, you will undervalue the amount of impact that your paid media is making.

Let’s say that you’re running an on-going top-of-the-funnel demand generation campaign and want to know the ROI. If you have a three month cycle from lead to close, you can’t use campaign data from the last three months. That’s because the demand that you generated hasn’t had enough time to turn into revenue yet. If you were to count the cost that generated the demand, but not give yourself a chance to count the revenue that you may get in the future, you’ll undervalue the ROI of your campaign.


paid media campaign


When calculating the ROI of your campaigns, you must understand the impact of your sales cycle. Use the sales cycle to figure out the lag between your action (engagement with your ads) and the result (revenue), and then remember to consider that time period when calculating ROI.

Mistake 3: Confusing what ROI means

This may seem like a strange mistake to make, but I’ve seen and heard of many marketers making the mistake of defining ROI as things other than revenue divided by cost. It’s fine to have other key metrics, but don’t call them ROI and don’t use them to replace true ROI when evaluating performance.

One metric that many paid media marketers use is Cost Per Deal. It’s a good metric to look at, but it can be misleading if you’re using that instead of ROI. The limitation of Cost Per Deal is that it doesn’t take into account deal size. If a campaign has a high cost per deal (maybe even above your target threshold), but is also bringing in bigger deals, it may be a good campaign to keep running.

Keep track of metrics like these, but at the end of the day, make decisions based on ROI (revenue divided by cost).

Mistake 4: Double-counting revenue

If you’re hacking an attribution solution, you may be at risk for double-counting revenue. This can happen two ways:

Calculating the ROI of marketing platforms separately. Let’s say a prospect clicks on your LinkedIn ad on Monday, an AdWords ad on Tuesday, and then converts on Wednesday. Somewhere down the line, she becomes a customer worth $20,000 in revenue. If you’re relying on the channels’ native analytics for attribution and ROI calculation, your reporting will look like this:

LinkedIn will say that it generated one conversion, which you will link to $20,000 in revenue. If you spent $10,000 on the campaign, you will calculate a LinkedIn ROI of 2x.

AdWords will also claim that it generated one conversion, which will also be linked to the $20,000 in revenue. If you also spent $10,000, you will calculate an AdWords ROI of 2x.

But when you do a calculation of your total marketing ROI, it will only be 1x -- $20,000 revenue from $20,000 in spend. The performance of both of your channels will be overstated.

Each channel doesn’t communicate with the other. AdWords isn’t telling LinkedIn that they also recorded a conversion, and therefore should split credit. That’s the role of a centralized attribution solution. It brings in all the data from each of your channels and applies credit according to your model, reconciling the disparate data.

Not weighting touchpoints properly. The second double-counting issue is that if you’re hacking multi-touch attribution by combining your disparate data into a spreadsheet and then connecting conversion data from the various channels to downstream revenue, it’s important to get your weighting calculations correct. This isn’t too much of a problem for smaller companies that may just be using a few campaigns on a few different platforms. But when you start running hundreds of campaigns on several different platforms, it’s easy to make small mistakes that lead to big impacts down the road.


double count


Accurately calculating ROI is a critical component of an effective paid media strategy. ROI should be the basis of decision-making, whether that’s doubling down on a great campaign or stopping an underperforming campaign altogether.

If you make one of these mistakes and pour money into a failing campaign, you’re throwing money away. But if you can calculate ROI confidently and accurately, you can make decisions that dramatically grow your impact on the business.