The purpose of measuring ROI is to know what works & what to do next.
ROI is a straightforward equation in theory with only two variables; How much did you spend? And, how much did you make?
If you know these values, you can use the formula above or a handy ROI calculator to measure your ROI.
In practice, understanding the causal relationship between these two variables (i.e. attribution) is hard.
So hard in fact that, despite a $1.8 billion marketing attribution industry, no one has truly solved ROI attribution.
What begins as an innocent attempt to improve budget allocation often descends into philosophical debates over attribution models, egos being bruised, and demoing the latest 7-figure Adobe product.
A More Pragmatic Approach
For most marketing teams, the relationship between confidence and value when measuring ROI follows a bell curve.
When you move from ad-hoc (non-existent) measurment of ROI on your marketing activity to a consistent (albeit imperfect) approach, you gain a lot of value and you’re able to make better, directionally correct decisions.
If you’re Coca Cola, Expedia, or have a marketing budget rivaling the GPD of a nation state – then yes, improving the accuracy of ROI calculations by a fraction is probably a good use of your budget.
While all businesses benefit from having a clearer idea of what’s working, the value of measuring ROI increases in proportion with the amount of money you’re spending.
Striving for perfect attribution is not just a fallacy but it ignores the fact that imperfect, yet directionally correct, ROI calculations can tell us what’s working and what to do next. For most businesses, this isn’t just acceptable – but optimal when considering the trade-offs of seeking perfection.
The Pragmatist’s approach to ROI
The pragmatic approach to measuring ROI is to use whichever metric is closest to revenue that you know to calculate the ROI of a campaign.
In other words, if you know how much revenue a campaign generated – great, use it! This is often quite rare, but you might know how many customers, leads or visits the campaign sent. In this case, you’d use the metric closest to revenue (i.e. revenue, then customers, then leads, then visits) to calculate a predicted ROI.
TrueNorth is able to go one step further by building a model of your funnel working out the rolling value of a visit, lead, or customer.
There are drawbacks to this, for sure.
If a campaign costing $800 generates 10 MQLs, which based on your MQLs being worth $145 each on average, your ROI would be predicted as $650 (81.3%). What if those 10 MQLs were worth $200 each? Or nothing?
This is part two of the attribution argument – and a valid one as there is money at stake.
However, given the alternatives, we believe the benefits of having a directionally correct method of measuring ROI that is transparent outweighs the drawbacks of either not measuring ROI, or only measuring it when the data is conveniantly available.