Every TV and radio campaign produces a paper trail: the record of what actually aired, when, and on which station. That record is the postlog, and it is the single most important input to measuring whether the campaign worked. It is also, week after week, the most tedious part of the job.
This guide explains what a postlog is, what makes postlogs hard to work with, how postlog analysis turns them into attribution, and the side benefits that come from actually reading them.
What a postlog is
A postlog (also called a post log, an as-run log, or an affidavit of performance) is the station's receipt for your buy: a line for every spot that ran, with the date and time it aired, the station or network, the market, the program or daypart, the spot length, the creative, and usually the rate.
It is the counterpart to the pre-log, the schedule of what was planned to run. The two are never identical. Spots get bumped, preempted, moved and made good, so the postlog is the version you measure against. Even then, postlogs are revised after the fact, so a week's file may change after you first receive it.
In practice everyone in the industry says postlog, spot or airing. Nobody says "as-run log" out loud.
Why postlogs are hard
If postlogs arrived as one clean, standard file, attribution would be easy. They do not.
- Every format imaginable. Excel workbooks, PDFs, CSV exports, portal downloads and emailed tables, each laid out differently, often differently from the same station the month before.
- Time zones. One file is in Eastern, another in the station's local time, a third has been adjusted to the market's time already. Get this wrong by an hour and every spot is matched to the wrong minute of response.
- Timing. Postlogs for the prior week usually land Tuesday to Thursday, which leaves very little time to read them before next week's buy has to be placed.
- Revisions and make-goods. Bonus airings, late runs and corrections trickle in after the first file.
- Volume. A schedule across dozens of stations in many markets is thousands of lines a month.
This is why large measurement vendors often expect the data to arrive on autopilot, and why so much of TV measurement quietly breaks at the intake step. The nuances have to be teased out by hand, at least the first time for each station.
How postlog analysis works
Postlog analysis is the process of turning those files into a single, clean record of airings and then measuring each one.
- Normalize. Read every file, whatever the format, into one table: station, market, date, airtime, length, creative, cost.
- Align time. Convert every airing to the time zone in which the response is measured, so the spot and the traffic it caused are on the same clock.
- Match to response. For each airing, pull minute-by-minute web traffic from Google Analytics 4 (read-only, nothing installed), plus calls or orders if you have them, for the market the spot aired in.
- Baseline and lift. Compare the response after the airing with what your own history says would normally have happened at that minute, on that day of the week. The difference is the spot's incremental lift. Windows close when traffic returns to normal, and spots that air close together share credit.
- Report. Hand the postlog back with new columns beside every line: incremental sessions, new users, calls or purchases, and cost per response; then roll it up by station, program, daypart, day of week, creative and market.
The output is the file you already have, made useful: a receipt that also says what each spot bought.
The side benefit: catching what the postlog gets wrong
Reading every airing's raw traffic does something the CPM never will. It audits the postlog.
- A flat line where a spike should be. When a logged spot shows no response at all, that is worth a call to the station. More than once the answer has been "it actually ran at a different time." The corrected airing then matches a spike that was sitting unexplained.
- A spike with no spot logged. Unexplained lifts often turn out to be make-goods or bonus airings that never made it onto the log. Now you know they ran, and what they were worth.
- Time-zone slips. A whole station whose spikes land an hour after its spots is a file in the wrong time zone, not a slow audience.
Each of these is a discrepancy between what you were billed for and what happened. Attribution is the reason to do postlog analysis; reconciliation comes free with it.
Software or done for you
There are two ways to run this.
- Software. Log in, connect GA4 read-only, upload your postlogs each week, and read the results. Clients on the other side of the world do this at 3 a.m. their time and never talk to anyone.
- Full service. Send the postlogs in whatever shape they arrive. They are transformed, uploaded, checked, and the report is delivered to your inbox, with an analyst on the phone for questions and custom cuts.
Either way, the first report is available the next business day after the postlog and GA4 access arrive.
Common questions
Can you work from a PDF postlog?
Yes. PDFs, Excel, CSV and portal exports are all normal.
Do I need to reformat the files first?
No. Send them as the stations send them.
How often should postlogs be processed?
Weekly, as they arrive, so the results are in hand before the next week's buy.
What if the postlog is revised after you have processed it?
Send the revision. The affected airings are re-run and the report is updated.
Does this work for radio postlogs?
Yes. Radio station logs and live-read times are processed the same way.
Getting started
Quality Analytics has processed postlogs and measured linear TV and radio response for advertisers and agencies since 2014, across hundreds of campaigns and hundreds of millions of dollars of media. Send us last week's postlogs and give read-only access to Google Analytics 4; you will be set up the same day and have your first report the next business day. We run up to a month of your airings before you pay anything, and plans start at $399 a month, month-to-month. Book a demo to see it on your own data.