Build an hours and days dashboard in Claude

Builds the Get MCP Ads Hours and days template as a Claude dashboard: spend, impressions, clicks, conversions and value by hour x weekday, by hour and by weekday on Meta, Google Ads, TikTok and Snapchat, by weekday on Pinterest and LinkedIn, with CPM, CTR, CPC, CPA, ROAS and conversion rate, averages per day that divide by how many of each weekday the period holds, and each hour's share of spend and conversions, against the comparison period the reader picks. Hours stay in each account's own time zone, today is left out, and conversions a platform does not report by hour stay blank.

Recipe

Intent A live page showing on which hours of the day and which days of the week the ads earn their spend, per platform, so the team can set ad schedules and bid adjustments and pick when to launch or change things.

Steps: 3
Tools: 1
Rules: 5
Failure modes: 3

skills/hours-and-days-dashboard.md

When to use it

Use it when

Someone asks when their ads perform best, which hours or days to cut or push, whether to set an ad schedule or dayparting, or for an hour by weekday heatmap they can keep in Claude.
Before changing an ad schedule or bid adjustments by hour or day, or choosing when to launch a campaign or change budgets, and to check afterwards what changed.

Do not use it when

Changing ad schedules or bids: the dashboard reads only; changes go through the platform tools with preview and confirmation.
Conversion timing on Meta, Snapchat or Pinterest by hour: those platforms do not report conversions by hour; their weekdays carry them.
Hosts without Claude's Dashboard artifact type: answer in the conversation with the matching reports instead.

What it needs first

Optional Meta Ads, Google Ads, TikTok Ads, Pinterest Ads, Snapchat Ads, LinkedIn Ads

No single source is required: this one reads across whatever you have connected.

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.

Evidence every conclusion must carry

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

Inputs
Input	Required	What it is
`window`		Analysis window.Default: last 30 days.

Procedure

In order. Every tool named is one the gateway ships, and links to its reference.

Steps

Resolve the client's workspace and its ad accounts, one per platform (Meta, Google Ads, TikTok, Pinterest, Snapchat, LinkedIn), with list_accounts.
Fetch the heatmap template with every account (accounts, one per platform, each with its currency).get_dashboard_template
Create a Dashboard artifact, write every document exactly as given in one batch, read its check, hand over with the lines given.

Decision rules

The observable condition, what it lets you conclude, and what takes the conclusion back.

When The template is refused with PLAN_REQUIREDConclude 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 which hours to cutConclude Compare each hour's share of spend with its share of conversions and its CPA within one platform, on enough days (four weeks at least), and say the hours are the account's own time zone; a cheap night hour can be cheap because few advertisers bid then, so propose a bid adjustment before a cut, and never compare hours across platforms.
When The person asks about conversions by hour on Meta, Snapchat or PinterestConclude Say Meta returns no website purchases with its hourly breakdown, Snapchat no pixel purchases by hour and Pinterest no hourly data past 8 days: judge those hours on delivery (CPM, CTR, CPC) and their conversions by weekday.
When One weekday stands outConclude Check its days in the day rows first: a single holiday or launch day can carry a weekday that holds only four or five occurrences in the period.
When A reader sees a source fail to loadConclude 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.

Where the agent stops

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

How it goes wrong quietly

The cases where the analysis is wrong and still looks right. Read them before trusting a number.

Editing the template's page or queries by hand loses the provenance of each figure: ask for changes by name instead.
Comparing weekday totals between ranges that hold a different mix of weekdays: compare averages per day, which divide by each weekday's own count.
Shifting one platform's hours into another's time zone, or adding conversions across platforms: each platform's hours and conversions are its own.