An automation you cannot audit is an automation you cannot trust. TrackIQ records **every run of every rule** — when it fired, what it matched, what it changed, and whether the change was applied — and keeps it as a browsable history.

This is the screen to open when a bid is not what you expected, or when someone asks what the rules have been doing.

[Watch the walkthrough](https://kb.trackiq.com/kb-assets/kbvideo-run-history-explainer-52c7f727.mp4)

*Prefer to watch? Eighty-five seconds on matched, changed, applied — and the rule that changes nothing.*

## **Last Run and Next Run, on the rules list**

![Last Run, Next Run, Run History and Changed on the rules table](/kb-assets/shot-b26-hist-view-651d68.png)

Scroll the Rules table right and four columns answer the day-to-day questions without opening anything:

| Column | Answers |
|---|---|
| **Last Run** | did it fire, and when? |
| **Next Run** | when will it fire again? |
| **Run History** | the icon that opens the full log |
| **Changed** | how many entities the last run touched |

**Changed is the column to scan.** Across a set of rules on the same schedule you will see numbers like 101, 13, 521, 8, 5, 162 — and the outliers are the story:

- A rule changing **hundreds** of entities every day is either managing a very large scope or has thresholds loose enough to catch everything. Check which.
- A rule changing **nothing**, run after run, is not protecting you. Its conditions never match, and it is giving you false confidence.
- A rule whose Changed count **jumps** is reacting to something that moved — a price change, a stock-out, a competitor. That jump is a signal about the account, not just the rule.

**Last Run not matching Next Run's cadence** is the first thing to check when a rule seems inactive. A rule showing a Last Run several days stale on a daily schedule did not run, and that is a different problem from a rule that ran and changed nothing.

## **The run history**

The **Run History** icon opens every run that rule has ever made.

![Every run, with its trigger, and what it matched, changed and applied](/kb-assets/shot-b26-runs-view-651d68.png)

> *Every run this rule has made. Click **View** to open a run and see exactly what it changed and why.*

| Column | What it tells you |
|---|---|
| **Run** | the run number, ascending over the rule's life |
| **When** | date and time it fired |
| **Status** | Completed, or otherwise |
| **Trigger** | **Scheduled · Auto Apply** or **Manual · Dry Run** |
| **Matched** | entities in scope that the rule evaluated |
| **Changed** | entities a branch decided to act on |
| **Applied** | changes actually written to Amazon |

## **Reading the three counts**

The relationship between Matched, Changed and Applied is where the useful information lives.

**Matched is your scope.** It should be roughly stable run to run — 266, 267, 268, 269. A drift means campaigns are entering or leaving the rule group. A sharp fall means something in scope was paused or removed.

**Changed ÷ Matched is how aggressive the rule is.** 133 of 266 means the rule is touching half of everything it looks at, every single day. On a daily schedule that is usually a sign the thresholds are too loose, or that the change cap is doing all the real work of keeping it safe.

**Applied should equal Changed on a scheduled run.** When it does not, changes were computed but not written — check Status, and check whether the Amazon connection is healthy. See **Checking When Your Data Last Synced**.

**Applied is blank on a dry run**, shown as a dash. That is the definition of a dry run rather than a fault.

## **Trigger tells you who did it**

The **Trigger** column separates the rule's own schedule from human intervention:

- **Scheduled · Auto Apply** — the rule fired on its cadence and wrote its changes.
- **Manual · Dry Run** — somebody previewed it. Nothing was written.

A **Run Now** from the rule's ⋮ menu also appears here, which matters during an incident: if a bid changed at an odd hour, the run history says whether the schedule did it or a person did.

## **Opening a run**

**View** opens the run and lists every entity it touched — campaign, ad group, target, which **branch** fired, and the value before and after.

That branch number is the part worth using. It ties the change back to a specific rule of yours, so "why did this bid drop?" resolves to "branch 3 matched, because orders were above 1 and ACoS was over target" rather than a guess. See [**Building a Rule from Scratch**](https://kb.trackiq.com/hc/knowledge-base/articles/building-a-rule-from-scratch) for how branch order decides which one wins.

## **The audit habit worth having**

**Weekly:** scan the Changed column across all rules. Anything at zero or in the hundreds deserves a look.

**After any threshold edit:** open the next run and check Changed moved the way you intended. This is the fastest feedback loop the product offers.

**When a number looks wrong:** open the run covering that date before assuming the rule is broken. Most surprises turn out to be a branch you forgot was there, firing exactly as written.

**Before pausing a rule:** check what it has been changing. A rule you no longer remember writing may be the only thing holding an account's ACoS down.

## **Related**

- [**The Change Log: Auditing What a Rule Changed**](https://kb.trackiq.com/hc/knowledge-base/articles/the-change-log) — the account-wide view across every rule.
- [**Dry-Run Preview**](https://kb.trackiq.com/hc/knowledge-base/articles/dry-run-preview) — producing a run that changes nothing.
- [**Look-back Windows and Adjustment Frequency**](https://kb.trackiq.com/hc/knowledge-base/articles/lookback-windows-and-adjustment-frequency) — what sets the cadence you are reading here.
- [**A Rule Didn't Run**](https://kb.trackiq.com/hc/knowledge-base/articles/a-rule-didnt-run) — when Last Run is stale.