How to Run a Matchback Report Without a Big Data Team
A plain walkthrough of matching marketing spend to named, closed sales, using tools an owner-operator already has.
TL;DR
A matchback report compares the list of people you marketed to against the list of people who actually bought. You can run a basic version with two spreadsheets and an email address or phone number as the match key. It tells you which campaigns produced named, closed sales instead of just clicks and opens.
A matchback report compares two lists: the people you marketed to and the people who bought. Where a name appears on both, you have a sale you can tie to a campaign. You can build a basic version with two exported spreadsheets and a shared match key like an email address or phone number. No data team required.
The point is simple. Activity reports tell you who clicked. A matchback tells you who clicked and then paid. For an owner-operator deciding where the next dollar goes, that difference is everything.
What is a matchback report, in plain terms?
A matchback takes your closed sales for a period and checks each one against the list of people you sent a campaign to. If a buyer was on the mailing list or the email send, the sale gets matched back to that campaign.
It works in the opposite direction from attribution tracking. Instead of following a click forward and hoping it ends in a sale, you start from confirmed sales and look backward to see who you had already reached.
That backward direction is why it holds up in the messy real world. People buy days or weeks later. They buy from a different device. They walk into a store after seeing a mailer. None of that breaks a matchback, because it only cares about two facts: did this person buy, and had you marketed to them. See Matchback Reporting for the longer explanation.
What do you need before you start?
Two exports and one match key. That is the whole list.
The first export is your campaign audience: every contact you sent the email or mailer to, with their email and, if you have it, phone and address.
The second export is your sales for the same window: every closed transaction with the buyer's contact details and the amount.
The match key is the field both lists share. Email is the cleanest. Phone is a solid second. Name plus address can work for direct mail, but typos and abbreviations make it messier.
How do you actually run the match?
Start by cleaning both lists so the match key looks identical on each side. Lowercase every email. Strip spaces. Reduce phone numbers to digits only. Small inconsistencies are the main reason real matches get missed.
Then line the two files up by the key. In a spreadsheet, a lookup formula handles this: for each sale, check whether the buyer's email appears in the campaign audience. If it does, mark the row as matched and note which campaign.
Total the matched rows. That sum is your campaign-driven revenue, meaning sales from people you had actually reached. Divide your campaign cost by the number of matched sales and you have a real cost per sale, not a cost per click.
If a buyer matches more than one campaign, pick a rule and stick to it. Crediting the most recent campaign before the purchase is common and defensible. The rule matters less than applying the same one every time.
Want this working on your numbers?
Viewmedia makes marketing you can prove, matched to real, closed sales.
How do you read the result honestly?
A match means the person bought and you had reached them. It does not prove the campaign caused the purchase. Some of those buyers would have bought anyway.
Treat the matched revenue as an upper bound on what the campaign influenced, not a precise causal figure. The useful comparison is across campaigns run the same way: if one mailer matches twice the sales of another at the same cost, that is a real signal worth acting on.
To get closer to cause, hold out a random slice of your audience and skip marketing to them. Compare the buy rate of the people you reached against the holdout. The gap gives you a cleaner read on what the campaign actually added. It is more work, but it turns a matchback into something close to a controlled test.
When is a manual matchback enough, and when is it not?
A spreadsheet matchback is enough when your volume is small, your data lives in one or two places, and you run it once a month. Most owner-operators are in exactly this position, and a simple monthly match beats any activity dashboard for deciding where to spend.
It breaks down when your buyer records are split across a point of sale, a booking system, and an email tool that share no common IDs. At that point reconciling the lists by hand eats more time than the report is worth, and small matching errors start distorting the numbers.
That is where a managed Provable Marketing setup earns its keep. It standardizes the match keys and runs the reconciliation for you. For consumer email specifically, Viewmedia guarantees a 15% open rate, and those sends feed straight into matchback so opens are never the end of the story. Unfamiliar terms are defined in the Glossary.
What mistakes throw off the numbers?
The most common is dirty match keys. A trailing space or a capital letter is enough to make a true match look like a miss, which quietly understates your results.
The second is mismatched date windows. If your sales export covers a different period than your campaign, the match is meaningless. Line the windows up before you run anything.
The third is claiming causation. A matchback shows correlation between reach and purchase. Report it that way, compare campaigns against each other, and run a holdout when you need a firmer answer.
Founder, Viewmedia
Brian Wroblewski is the founder of Viewmedia. For more than two decades he has helped local and regional businesses turn marketing spend into provable, closed sales.
