# Build a change impact dashboard in Claude

> Tier: recipe. Operating recipe: real tool sequence, core rules, known traps.

Builds the Get MCP Ads Change impact template as a Claude dashboard for Meta and Google Ads: every change of the period from Meta's activity log and Google Ads' change history (budgets, bids and bid strategies, statuses, targeting, creatives, ads and campaigns launched) with who made it, its old and new value and, for changes made through Get MCP Ads, the reason given; for each, the changed object's spend, conversions, value, CPM, CTR, CPC, CPA and ROAS over the same number of whole days before and after, next to the rest of the account over those days, with overlapping changes and low volume flagged; changes by type, the account by day with its changes, and the team's events from the client journal.

**Intent.** A live page showing what changed in the ad accounts and what the changed campaign, ad set or ad did afterwards, so the team can tell which changes went with better or worse results and what to roll back or repeat.

## When to use

- Someone asks what changed in an account, why results moved after a change, whether a budget increase, a bid strategy switch or a new ad helped or hurt, or for a change log with its impact they can keep in Claude.
- A weekly review of the changes the team and the platforms made, before deciding what to roll back or repeat.

## When not to use

- Rolling a change back: the dashboard reads only; changes go through the platform tools with preview and confirmation.
- TikTok, Pinterest, Snapchat or LinkedIn changes: none offers an account change history the dashboard can read; say so and use the client journal for what went through Get MCP Ads.
- Hosts without Claude's Dashboard artifact type: answer in the conversation with the matching reports instead.

## Sources

Required: any connected source
Optional: Meta Ads, Google Ads

## Preconditions

- The organisation is on Team or Enterprise: ready-made dashboards are not offered on Free or Pro.
- The account is among the chosen accounts, confirmed via list_accounts.

## Inputs

- `window`: Analysis window. Default: last 30 days.

## Procedure

1. Resolve the client's workspace and its Meta and Google Ads accounts with list_accounts.
2. Fetch the change-impact template with those accounts (accounts, one per platform, each with its currency). — tools: `get_dashboard_template`
3. Create a Dashboard artifact, write every document exactly as given in one batch, read its check, hand over with the lines given.

## Decision rules

- **When** The template is refused with PLAN_REQUIRED → **conclude** Say the dashboards come with Team and Enterprise and that only the account owner changes the plan; answer the question in the conversation meanwhile.
- **When** The person asks whether a change worked → **conclude** Read the change's before and after against the rest of the account over the same days, say how many days and conversions each side holds, and call it an observed difference, not proof: a change flagged overlap or low_volume gets no verdict.
- **When** The person asks about changes older than 30 days on Google Ads → **conclude** Say Google keeps 30 days of change history: older changes come from the client journal's daily archive, without old and new values, and only for accounts the archive followed then.
- **When** A change shows no_after or no_before → **conclude** A pause, a removal or a launch leaves no side to compare on the object itself: read its campaign or ad set and the rest of the account instead.
- **When** A reader sees a source fail to load → **conclude** Their own Get MCP Ads access decides: check that their connection covers this client's workspace and that their plan includes dashboards before changing the dashboard.

## Evidence requirements

- Every figure carries its account, metric and window.
- Anything below a readable sample size is reported as unjudged, not as zero.

## Human approval boundary

- The readout changes nothing. Any change it motivates goes through Safe Writes: preview, human confirmation, then apply.

## Failure modes

- Editing the template's page or queries by hand loses the provenance of each figure: ask for changes by name instead.
- Crediting a change with a move the whole account made: the rest of the account over the same days is the baseline.
- Giving an effect to one of several changes made together on the same object, its campaign or ad set, or the account.

## Output

- **Readout**: The figures, with windows and accounts named.
- **Flags**: What deserves a deeper skill or a human decision.

## Related

- Skill: [scorecard-dashboard](https://www.getmcpads.com/skills/scorecard-dashboard.md)
- Skill: [budget-pacing-dashboard](https://www.getmcpads.com/skills/budget-pacing-dashboard.md)
- Skill: [creative-analytics-dashboard](https://www.getmcpads.com/skills/creative-analytics-dashboard.md)
- Skill: [hours-and-days-dashboard](https://www.getmcpads.com/skills/hours-and-days-dashboard.md)