A reliable PPC account change log turns scattered edits into an evidence trail. This guide provides a reusable spreadsheet structure, platform-specific recording guidance for Google Ads, Microsoft Ads, and Meta Ads, plus a simple method for estimating whether a change improved performance or should be rolled back.
Overview
Paid media performance rarely changes for one reason. A new bid strategy, budget adjustment, audience edit, tracking update, creative refresh, or landing page change may overlap with seasonality, promotions, auction conditions, and conversion delays. Without a documented account history, it is easy to mistake correlation for causation or forget which version was active during a reporting period.
A PPC account change log creates a shared audit trail. It should answer five practical questions:
- What changed, and where?
- Who made the change and when?
- What was the reason or hypothesis?
- What outcome was expected, and over what evaluation window?
- Did the team keep, modify, or roll back the change?
Use one row for each meaningful change rather than one row for an entire work session. Grouping unrelated edits makes later analysis difficult. A single row may cover a coordinated experiment, such as changing the target CPA and reallocating budget at the same time, but the relationship should be explicit.
This process supports campaign optimization, paid search analytics, paid social analytics, and stakeholder reporting. It also complements a cross-channel paid media reporting dashboard by explaining why the numbers moved.
How to estimate
The log is most useful when each entry includes a simple estimate of expected impact. The estimate does not need to be precise; its purpose is to make the original decision testable.
Use a repeatable impact model
For a search or social campaign, start with the following baseline relationships:
- Estimated conversions = expected clicks × expected conversion rate
- Estimated cost per conversion = expected spend ÷ estimated conversions
- Estimated revenue = estimated conversions × expected value per conversion
- Estimated ROAS = estimated revenue ÷ expected spend
Record the baseline and expected values in separate columns. For example, if a landing page change is expected to increase conversion rate while click volume remains broadly stable, document the current conversion rate, the assumed new rate, the traffic window, and the reason for the assumption. Do not present the estimate as a forecast with guaranteed accuracy.
Define the evaluation window before the edit
Choose a review period appropriate to the campaign's volume and conversion delay. A high-volume campaign may produce an initial directional signal quickly, while a lower-volume campaign may require more time. Note the planned start and end dates, and state whether the evaluation uses click date, conversion date, or the platform's reporting view.
Keep the comparison consistent. Compare the same campaigns, conversion actions, attribution setting, geography, device grouping, and date logic where possible. If those inputs change, record the difference rather than treating the result as a clean before-and-after test.
Use a decision rule
Before reviewing results, define what would lead to each decision:
- Keep: the primary metric improves without an unacceptable deterioration in a guardrail metric.
- Modify: the direction is promising, but the result is mixed, incomplete, or operationally inefficient.
- Roll back: the change causes a material deterioration, breaks tracking, creates delivery problems, or fails a predefined business constraint.
- Inconclusive: the evaluation window or volume is insufficient to make a responsible decision.
Inputs and assumptions
Copy the following columns into a spreadsheet, database, or campaign reporting workflow. Keep the field names stable so entries remain comparable over time.
| Field | What to record |
|---|---|
| Change ID | A unique reference such as 2025-014. |
| Timestamp and time zone | When the edit was made, including the account's reporting time zone. |
| Editor | User, team member, automation, script, or integration responsible for the edit. |
| Platform and account | Google Ads, Microsoft Ads, Meta Ads, or another connected advertising platform. |
| Campaign and object | Campaign, ad group or ad set, keyword, audience, asset, conversion action, or account setting affected. |
| Change type | Budget, bid, targeting, keyword, creative, tracking, landing page, schedule, or structural change. |
| Before and after | The prior value and the new value, using exact setting names where possible. |
| Reason and hypothesis | Why the edit was made and what outcome is expected. |
| Baseline metrics | Recent spend, impressions, clicks, conversions, conversion rate, cost per conversion, revenue, or ROAS. |
| Evaluation window | Review start date, end date, conversion lag consideration, and comparison period. |
| Result and decision | Observed outcome, confidence level, keep/modify/rollback/inconclusive decision, and follow-up owner. |
Platform-specific recording guidance
| Platform | Useful change-log practice |
|---|---|
| Google Ads | Use the account's change history to verify the editor, timestamp, affected campaign or asset, and recorded before-and-after details. Supplement the platform record with the business reason, hypothesis, evaluation dates, and rollback plan. |
| Microsoft Ads | Record the campaign, ad group, keyword, targeting, budget, bid, and import-related context. If an edit follows a Google Ads import, note what was adapted rather than assuming settings transferred exactly. See the Google Ads to Microsoft Ads import checklist. |
| Meta Ads | Identify the campaign, ad set, ad, audience, budget, optimization event, creative, or placement affected. Distinguish a manual edit from an automated rule or bulk change, and save the relevant version or export when available. |
Platform history and an internal change log serve different purposes. Platform history is the system record of what happened. The internal log explains why it happened, what the team expected, and what decision followed. Check the current interface, export fields, API availability, and automation labels before standardizing a workflow. Revisit those details at least twice yearly or after a major platform or automation update.
Worked examples
Example 1: Search bid strategy change
A search campaign changes from one bidding approach to another to improve efficiency. The log should include the exact prior and new setting, the affected campaigns, the date and time, and whether the change was made manually or through an automated workflow. The hypothesis might be: “The new bid strategy should reduce cost per conversion while maintaining conversion volume.”
Assume the baseline is 120 conversions from $6,000 in spend, or $50 per conversion. The working estimate is 110 conversions from $4,950, or $45 per conversion. The evaluation window is recorded in advance, along with guardrails such as a minimum acceptable conversion volume and a maximum acceptable cost per conversion. If the result is 108 conversions from $4,860, the observed cost per conversion is also $45. The entry can be marked “keep,” subject to the campaign's business goals and any conversion-delay considerations.
Example 2: Budget reallocation
A publisher moves budget from a campaign with weaker marginal returns to one with stronger recent performance. Record the amount or percentage moved, the source and destination campaigns, the assumptions behind the decision, and the date the new budgets took effect. Include a follow-up date so the change is not forgotten during the monthly review. For a deeper allocation process, see Marginal ROAS Budget Reallocation.
Example 3: Tracking or landing page change
When a conversion tag, UTM structure, form, or landing page changes, mark the entry as a measurement-risk change. Record the old and new conversion action or URL, test status, and validation owner. Do not judge the campaign solely on reported conversion volume until tracking has been checked. A conversion tracking audit and the paid media attribution reconciliation workflow can help separate a real performance shift from a reporting issue.
When to recalculate
Update the estimate and review the log whenever a material assumption changes. Recalculate after a budget, bid, targeting, keyword, creative, conversion tracking, landing page, attribution, or automation edit. Also revisit entries when pricing inputs, conversion values, traffic mix, seasonality, or business priorities move.
Use the log during three recurring reviews:
- Weekly: scan recent edits for broken tracking, unusual spend, delivery changes, or overlapping tests.
- Monthly: connect changes to campaign performance and mark each entry as keep, modify, roll back, or inconclusive.
- Quarterly or after major platform changes: verify platform change-history fields, export options, API or script workflows, campaign setting names, and permissions.
Before making the next significant edit, duplicate the latest row structure, capture the baseline, write the hypothesis in one sentence, and define the decision rule. Then link the entry to the relevant creative, keyword, landing page, or budget analysis. For example, a keyword-related change can be paired with a negative keyword conflict check, while a creative change can reference the paid ad creative refresh calendar.
A good PPC account change log is not extra reporting for its own sake. It is a lightweight control system: document the decision, measure against the stated expectation, and preserve a clear path back when an edit does not work.