I'm doing a heap of email marketing recently. It's hard to attribute but here's how I prove ROI.

Email is notoriously hard to attribute. Sure you can look I clicked data, but that doesn't really tell us anything other than this got clicked.

When you are growing a SaaS company, you need to be able to show which activities are actually leading to revenue.

And simply using UTM links is not enough to show that for most SaaS companies.

Why?

Because frankly, unless that person converts on the exact same session where they click the link, you're really heavily relying on an attribution model to do the hard work for you.

A method that I have used in multiple different companies is to look at correlation between days when emails were sent and revenue bumps.

For example, on a recent client, I discovered that there was a 70% likelihood of an above-average day in terms of revenue whenever we sent an email.

The click data can't tell us that kind of thing.

So here's how it works:

Step 1

Open up a new Google Spreadsheet and create a tab with the export of your MRR data from the past 90 days.

If you use a tool like ChartMogul or ProfitWell, you should be able to get this very easily.

I recommend that you delete the churn and contraction columns of your MRR data. All you need is:

• New business MRR

• Reactivation MRR

• Expansion MRR

Effectively, all of the positive growth.

Step 2

Now go to your email marketing tool. Open up the broadcast tabs for the same period of the revenue data that you're looking at.

Export the emails that you sent with the dates that they were sent on. Don't worry about duplicates sent on the same date, it doesn't matter at this phase.

You're also not looking at lifecycle emails here, only broadcast.

Step 3

Add on a new Spreadsheet tab the export of emails with dates that they were sent.

Step 4

Create a new column on your MRR data spreadsheet. Call it 'email sent?'. Now do a COUNTIF formula that queries whether there was an email sent on the same day as you're currently looking at. Here's a sample formul where I have a sheet called 'Email Data' and column B contains the dates emails were sent, and a MRR data sheet where cell A2 contains a date with revenue data:

=COUNTIF('Email Data'!B:B, A2) > 0

Step 5

Now create a column that calculates the additional revenue for each day. In my example above, that contains data from column B, column E, and column H.

Step 6

Now add another column that calculates the average additional revenue on days that emails were not sent. This will give us a baseline to explain how much revenue we might normally expect on any given day.

Step 7

Create a formula which counts the number of days that were above baseline when an email was sent vs. not sent. This will result in two different numbers.

Then create a formula that just counts up all of the number of days where an email was sent vs not.

This will allow you to create a percentage of days where the baseline was crossed by either sending or not sending an email.

Here's the little comparative section of my spreadsheet looks like: