Ad Budget Pacing
Maya-Beth FinottiTracks campaign or account spend against a period budget across ad platforms, flags meaningful under- or over-pacing, and calculates the daily correction required.
By agent · Claude Code
Skills in the open SKILL.md format that Claude Code reads. Each one has a plain-language summary and the command to install it.
$ npx skills add <author>/<skill>Replace the placeholder with the author and skill name from the skill's page. Start a new Claude Code session afterwards so the skill is picked up.
Tracks campaign or account spend against a period budget across ad platforms, flags meaningful under- or over-pacing, and calculates the daily correction required.
Routes paid advertising projects to the right skill in a 31-skill collection, preserving shared context across sessions and identifying gaps when no route fits.
Diagnoses underperforming paid ad accounts by tracing tracking, structure, targeting, creative, bidding, offer and external causes.
Sets paid-media CAC, ROAS, cash and kill-switch guardrails from contribution margin, payback and runway, with owners, overrides and restart rules.
Diagnoses whether a running ad is suffering creative fatigue or a confounder across Meta, Google Ads, LinkedIn and TikTok, then recommends the next action.
Computes CAC, ROAS, MER and payback from advertising spend data, then judges spend health against break-even, history and external benchmarks.
Maps an ICP and buying signals into sized, budgeted ad audience tiers for prospecting, lookalikes, first-party lists and retargeting without overlap.
Allocates a fixed paid-media budget across campaigns, platforms and funnel stages using marginal return, evidence gates and bounded reallocation rules.
Plans budget increases for proven paid campaigns, choosing step sizes, hold periods and rollback triggers while checking marginal economics, evidence and capacity.
Maps B2B buying committees and multi-decider household purchases to role-specific messaging, ad-platform targeting proxies and buying-stage sequences.
Selects paid advertising channel families by business economics, audience, funnel stage, budget and allowable CAC, then defines a focused test plan.
Plans paid-advertising careers with interview preparation, skill-gap roadmaps, NDA-safe portfolios and choices between agency, in-house and freelance work.
Builds hiring packets for paid-advertising roles, covering role scope, scorecards, interview loops, work samples and staged 30-60-90 ramps.
Designs multi-stage retargeting plans for paid advertising using conversion data, audience exclusions, message ladders, frequency caps and measurement tests.
Documents Browser Use Cloud APIs and SDKs for hosted browser automation, sessions, profiles, CDP connections, proxies and integrations.
Creates an end-to-end map of a customer experience, covering stages, touchpoints, emotions, pain points and improvement opportunities.
Injects external Dart files into dartdoc documentation, with regions and hidden setup lines for focused, analyzable code examples.
Makes .NET projects compatible with Native AOT and trimming by resolving IL analyzer warnings with annotations, refactoring and source-generated JSON.
Creates and reviews .NET P/Invoke and LibraryImport declarations for native libraries, including types, strings, memory ownership and handles.
Scaffolds new agent skills for the dotnet/skills repository, including frontmatter, directory structure, workflow guidance and validation.
Replaces fragile Dart path string operations with cross-platform handling from package:path and package:file.
Migrates existing C# projects to nullable reference types, resolving compiler warnings while preserving runtime behaviour and API intent.
Imports projects and their Git history into an Nx workspace, adapting paths, dependencies, plugins and TypeScript settings for monorepo use.
Provides language-specific testing guidance for the .NET code-testing pipeline, including commands, conventions and common errors.
Questions
A folder with a SKILL.md file: instructions, and sometimes scripts and reference files, that Claude Code loads when a task calls for them.
Each skill page has an install command. Run it in a terminal, then start a new Claude Code session so the skill is picked up.
An instructions file is loaded in every session and describes the project. A skill is loaded only when a task calls for it, so it can be long and specific without costing context on every turn.