Pacing
by
Daily spend pacing for ecommerce teams: read where the month actually stands against plan, then adjust today's spend by the amount that closes the real gap.
Pacing decisions get made every morning on yesterday's dashboard, which means they are made on one channel's version of revenue and without knowing whether the month is genuinely behind or simply weighted differently this year. The usual correction is to push spend up, which lands in the channels that report fastest rather than the ones with room. This skill reconciles spend and revenue to date against plan on one definition, separates a real gap from normal within-month shape, and returns today's adjustment with the channel it should come from.
What It Does
- Reconciles spend and revenue month to date on one definition across channels.
- Compares progress against the plan's own within-month shape, not a straight line.
- Separates a genuine shortfall from a month that is simply weighted later.
- Computes the daily spend needed to close the gap by month end.
- Identifies which channels have room to absorb the change efficiently.
- Returns today's adjustment, up or down, with the channel split.
Use Cases
- Stop correcting for a gap that is just the shape of the month.
- Push the increase into the channel with room rather than the one reporting fastest.
- Catch a genuine shortfall in week two rather than in the last week.
- Pace down deliberately when the month is ahead instead of spending the surplus.
Example Session
Are we on pace this month, and what should today's spend be?
Day 14 of 30. You are 6% behind plan, and about half of that is shape rather than shortfall.
- Line | Actual | Plan to date | Gap
- Net revenue | $312k | $332k | Minus 6%
- Expected from shape | | $320k | Minus 2.5% real
- Spend to date | $84k | $81k | Plus 4%
- Blended CAC | $44 | $39 | Plus 13%
The real gap is 2.5%, not 6%: last year's month was back-weighted the same way and the plan follows that shape. Spending is already 4% ahead at a CAC 13% above plan, so pushing more budget in makes the month worse, not better. Hold total spend flat today and move $1,200 from Meta prospecting into non-brand shopping, which is the only channel currently under its CAC target.
Why This Beats Prompting It Yourself
A Straight Line vs The Month's Real Shape: The Hidden Risk
A generic prompt divides the monthly target by the days and compares. Almost no ecommerce month is shaped that way: paydays, promos, weekends and seasonality all weight it, and a straight line reports a shortfall in the first half of nearly every month. The correction that follows pushes spend up at exactly the point in the month when it is least efficient.
When You Still Need To Know Where The Money Should Come From
An adjustment is only useful if it names the channel. This skill states a 6% headline gap that is 2.5% once shape is accounted for, against spend already 4% ahead at a CAC 13% above target, and concludes that holding is correct. The reallocation it does recommend goes to the one channel under its CAC target, which is a decision rather than a reflex.
Known Limitations
Within-month shape is learned from your own history, so a business under two years old or one that has changed its promo calendar will get a wider band and the output says so. Same-day revenue is always incomplete, which the skill accounts for using your own reporting lag rather than assuming the day is closed. It recommends the adjustment and does not change budgets in the ad accounts.
Install It
- Download the ZIP.
It is free and there is no account to create. - Unzip it into your agent's skills folder.
Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead. - Ask your agent to use it.
Restart the agent if it was already running, then it picks the skill up with no config.
skills folder
Copy
~/.claude/skills/pacing/
SKILL.md
references/month-shape.md
pacing/
# one file per day, with the adjustment
# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills
# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\Frequently asked questions
How is this different from a pacing dashboard?
A dashboard compares actual to a straight line and shows a gap. This compares against the month's own historical shape, separates the real shortfall from the expected weighting, and returns an adjustment with the channel it should come from. Most mornings the honest answer is to hold.
Which agents does the skill run in?
Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.
Why does it so often recommend holding?
Because most apparent gaps in the first half of a month are shape, and spending into them raises CAC without moving revenue. Holding is a decision here, supported by the shape figure, rather than an absence of one.
What does it need connected to work?
Spend across channels, orders and revenue, and the plan itself. Two years of history makes the shape model reliable. With less it still runs, widens the band, and says how much of the gap it cannot confidently classify.
How does it handle today's incomplete data?
By applying your own observed reporting lag per channel rather than treating the day as closed. Platforms report at different speeds, and a pacing call made on a partial day without adjusting for that is wrong in a predictable direction.
What is included with the skill?
The skill itself, the month shape reference, the daily pacing template, and the reporting lag profiles per channel. It is free and security scanned. We re-publish the ZIP when the shape model changes, so download it again if your pacing calls start drifting.
Today's pacing depends on where the month actually stands.
Polar reconciles spend and revenue to date against plan on one definition, so the adjustment answers the real gap and not yesterday's dashboard.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from a pacing dashboard?","acceptedAnswer":{"@type":"Answer","text":"A dashboard compares actual to a straight line and shows a gap. This compares against the month's own historical shape, separates the real shortfall from the expected weighting, and returns an adjustment with the channel it should come from. Most mornings the honest answer is to hold."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"Why does it so often recommend holding?","acceptedAnswer":{"@type":"Answer","text":"Because most apparent gaps in the first half of a month are shape, and spending into them raises CAC without moving revenue. Holding is a decision here, supported by the shape figure, rather than an absence of one."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"Spend across channels, orders and revenue, and the plan itself. Two years of history makes the shape model reliable. With less it still runs, widens the band, and says how much of the gap it cannot confidently classify."}},{"@type":"Question","name":"How does it handle today's incomplete data?","acceptedAnswer":{"@type":"Answer","text":"By applying your own observed reporting lag per channel rather than treating the day as closed. Platforms report at different speeds, and a pacing call made on a partial day without adjusting for that is wrong in a predictable direction."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the month shape reference, the daily pacing template, and the reporting lag profiles per channel. It is free and security scanned. We re-publish the ZIP when the shape model changes, so download it again if your pacing calls start drifting."}}]}</script>