The changelog is the asset, not the output: An agent's value compounds only when what it did is recorded in a form a later causal read can actually use. The changelog outlives both the agent and the vendor.
Read the full article below for detailed insights and actionable strategies.
Attribution by the numbers
iOS tracking loss
Google Brand cannibalization
Klaviyo overstatement
TikTok attribution lag
The listings are replaceable. The record of what changed and when is not. That record is what any later measurement depends on.
Why the record is the durable part
A batch of rewritten listings is worth whatever it earns this quarter. The log of which products changed, when, and what was left alone is what lets you answer the same question next quarter and the one after.
Swap the agent, swap the vendor, swap the platform: the log still works, because it describes your catalogue rather than their tooling.
What a usable log contains
| Field | Used for |
|---|---|
| Batch timestamp, to the hour | Before and after boundary |
| Product ids changed | Defining the treated set |
| Product ids deliberately excluded | Defining the control |
| Field-level description of the change | Knowing which lever moved |
| External events in the window | Ruling out confounds |
Field-level matters. "Updated listings" is not a lever. "Rewrote the first paragraph of the description, left titles and images alone" is.
Where teams lose it
The log lives in the agent's transcript, the transcript lives in a session, and the session is gone in a month. Or it lives in a chat thread nobody exports. Either way the information existed and was not kept.
Write it to something you own, at the moment of the change. A row in your own database beats a perfect account that has been deleted.
The connection to ROAS
Without the log, every later read is a before-and-after across a period where an unknown set of things changed, which is the weakest design there is. With it, the same read has defined boundaries and a control group, and returns an estimate with a confidence interval instead of a difference of two numbers.
That is the whole gap between a number you can act on and a number you can only quote.
The read
A causal read on a Google Analytics export, with your changelog supplying the boundaries, is 99 euro once, refunded if it does not move a budget decision. The interactive demo shows the fields with no signup.
Related answers
Key Terms in This Article
Analytics
Analytics is the systematic computational analysis of data. It reveals customer behavior and measures campaign performance.
Attribution
Attribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
Causality
Causality is the relationship where one event directly causes another, essential for identifying specific actions that drive desired outcomes in marketing.
Confidence Interval
Confidence Interval is a statistical range of values that likely contains the true value of a metric. In marketing analytics, it quantifies uncertainty around estimates, indicating the precision of an outcome or causal effect.
Control Group
Control Group is a segment of an audience intentionally not exposed to a marketing campaign, used to measure the campaign's true causal impact.
Google Analytics
Google Analytics is a web analytics service that tracks and reports website traffic.
Incrementality
Incrementality measures the true causal impact of a marketing campaign. It quantifies the additional conversions or revenue directly from that activity.
Session
A Session is a group of user interactions with your website within a given timeframe. It can include multiple page views, events, and transactions.
Related Articles
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Ready to see your real numbers?
Own the budget? Upload your GA4 export and see which channels drive incremental sales, with confidence intervals, in minutes. Have to defend it? Start with the live demo and take the read to your CFO.
Full refund if you don't see value.
Stay ahead of the attribution curve
Weekly insights on marketing attribution, incrementality testing, and data-driven growth. Written for the person who owns the budget and the person who has to defend it.
No spam. Unsubscribe anytime. We respect your data.
Frequently Asked Questions
Why is the changelog more valuable than the changes?
The listings earn what they earn this quarter. The record of what changed, when, and what was left alone is what lets you answer the same question in later quarters, and it survives changing agent, vendor or platform.
What level of detail does a changelog need?
Field level. Updated listings is not a lever. Rewrote the first paragraph of the description and left titles and images alone is, because it names which thing could have caused the effect.
Where should the changelog live?
Somewhere you own, written at the moment of the change. Agent transcripts and chat threads disappear, and a perfect record that has been deleted is worth less than a rough row in your own database.