Skip to content

Makes

LinkedIn Ads plan

Plans LinkedIn Ads audiences, the brand and demand split, formats and Thought Leader Ads, sized to how many of your target accounts are buying now.

Other tools

Add this skill to Claude

Claude takes skills as a ZIP file. Your download has started; if it didn't, use the button below.

  1. 1In Claude, open Customize, then Skills.
  2. 2Choose +, then Upload a skill, and pick linkedin-plan.zip.
  3. 3Ask for it by name, or let Claude use it when a task matches.

Skills need code execution turned on in Claude's settings. On a Team or Enterprise plan, an admin can share it with everyone.

Download the folder

Try it now, without installing

See the skill work on a made-up input before you add it. Nothing is installed, and you can swap in your own input.

Open in the Claude app

Opens a new chat in the Claude desktop app with the prompt filled in. Nothing is sent until you press enter.

Includes the whole SKILL.md and the sample input. Paste it into Claude, ChatGPT or any chat.

Influenced by
The 95-5 rule, Thought Leader Ads
Needs
Your target account list and buying group roles · The most a lead may cost
Works with
Claude, ChatGPT, Codex, Cursor and any agent that reads SKILL.md
Version
1.0.0 · Last checked · License MIT
Maintained by
Sabih AhmedPaid media
SkillPaid media

LinkedIn Ads plan

Example output

Plan LinkedIn Ads for a quarter
LayerBudgetCall
Demand$5,775Under ceiling
Brand$12,9604 a month
Unplanned$3,265Hold

Plans LinkedIn Ads audiences, the brand and demand split, formats and Thought Leader Ads, sized to how many of your target accounts are buying now.

The skill

The whole skill, in plain text.

Nothing is hidden. Read it, change it to fit your team, or take the whole folder.

Scripts
Optional Python scripts for the counting. It works without them and says which way it ran.
Network
Doesn't need the internet.

3 test cases

linkedin-plan/SKILL.md

In the folder

  • linkedin-plan/SKILL.md
  • linkedin-plan/references/linkedin-specs.md
  • linkedin-plan/scripts/linkedin_budget.py
---name: linkedin-plandescription: >-  Plans LinkedIn Ads for a B2B SaaS company: the target accounts and buying group roles, how many of those accounts  are in market this quarter, the split between a brand layer for the accounts that aren't buying yet and a demand  layer for the ones that are, the formats for each (Thought Leader Ads included), the settings, and the budget, all  held to the company's own cost per lead ceiling. Checks the plan and returns READY, READY WITH GAPS, NOT YET or  BLOCKED. Use it before launching or rebuilding LinkedIn Ads. Not for writing the ads (use ad-themes, then  ad-variants) or for deciding whether LinkedIn should get budget at all (use channel-fit).license: MITcompatibility: >-  Works in Claude, ChatGPT, Codex, Cursor and any agent that reads SKILL.md. Uses a connected ad account or CRM if the  agent has one, otherwise Campaign Manager forecasts and exports the reader pastes. Python 3 is optional, for the  sizing.metadata:  author: SHAKEUP  version: "1.0.0"  stage: make  plugin: paid-media  homepage: https://www.shakeuphq.com/skills/linkedin-plan--- # LinkedIn Ads plan Sizes LinkedIn Ads to how many target accounts are buying now, and keeps the rest hearing from you until they are. ## Purpose - One job: produce one LinkedIn Ads plan (audiences, layers, formats, settings, budget, measurement) and check it.- Stage: make. It ends with its own check, at most 2 fix rounds.- It never invents an audience size, a CPM, a CPC or a conversion rate. Those come from the reader's Campaign Manager forecast or past campaigns, or they are `not available`.- It never launches, edits or funds a campaign. It recommends; the reader applies. ## Use when / not | Situation | Use || --- | --- || Launching LinkedIn Ads, or rebuilding an account that only runs lead gen | this skill || Deciding the brand and demand split on LinkedIn | this skill || Choosing which founder or employee posts to sponsor | `founder-to-ads`, after this skill sets the Thought Leader Ads slot || Writing the themes and the ad copy | `ad-themes`, then `ad-variants` || You don't know the most a lead may cost | `lead-value-math` first || Deciding whether LinkedIn should get budget at all | `channel-fit` | ## Inputs | Input | Required | Ways to provide, in order | If missing || --- | --- | --- | --- || Target account list (count, and the list itself for upload) | yes | the CRM (accounts tagged as target); a CSV; the reader's count | ask once; `BLOCKED` without it or an ideal-customer description that LinkedIn targeting can express || Buying group roles | yes | `### Buying group` in the profile | ask once for the roles and who decides || Max CPL | yes | `lead-value-math` handoff; the reader | run without the ceiling check, coverage `partial`, and recommend `lead-value-math` || Years between purchases in the category (how often an account buys or replaces this kind of tool) | no | the reader's renewal and switching data; the reader's estimate (labelled) | in-market share `not available`; the split comes from the budget, not from the market || Lead target, or pipeline target with LinkedIn's share, ACV and lead-to-opportunity rate | no | `## Targets`; `## Company`; the reader | demand layer sized from the budget only || CPC and click-to-lead rate, CPM | no | the reader's own LinkedIn campaigns; Campaign Manager's forecast for the planned audience | that layer's budget `not available`; never a benchmark || Audience size for the roles at the target accounts | no | Campaign Manager's audience forecast after targeting is set | brand budget `not available` || Reach share and frequency per month the reader wants for the brand layer | no | the reader | ask once; the plan shows the cost per extra point of frequency instead || Monthly LinkedIn budget | no | `## Channels`; the reader | the plan states what it needs instead of fitting a number || Posts that could run as Thought Leader Ads | no | the reader; `founder-to-ads` handoff | Thought Leader Ads slot left open | ## Profile read From `company-profile.md`: `## Company` (ACV, sales cycle, regions), `## Buyers` (ideal customer, bad fit, buying group), `## Buying moments` (for which roles to reach and when), `## Messaging house` (pillars the layers carry), `## Claims and proof` (for any post or ad text), `## Channels` (LinkedIn budget, owner), `## Targets`, `## Accounts and data` (CRM, whether LinkedIn's Conversions API can receive CRM stages), and every `## Learnings` entry for `linkedin-plan` with status `noted`, `fact` or `accepted`. No profile: ask in one message for the target account count, the buying group roles, the monthly budget and the most a lead may cost. Say that `company-profile` saves this next time. ## Steps 1. **Apply learnings.** Past results (a format that produced opportunities, a role that never converted) change the defaults. Name each one that fires. 2. **Define the audience.** Target accounts uploaded as a company list. Roles from the buying group, expressed as LinkedIn job functions, seniorities or titles. One audience per layer, not per role, unless the message differs by role (an audience that would get the same ad is one audience). 3. **Size the market.** In-market share per quarter = 1 / (4 x years between purchases). In-market accounts = target accounts x that share. This is the 95:5 logic in `references/linkedin-specs.md`: use the reader's cycle, never the 5% figure as a default. 4. **Size the demand layer** (accounts buying now: lead gen and retargeting).   - Opportunities needed per quarter = pipeline target x LinkedIn's share / ACV. Leads needed per month = opportunities / lead-to-opportunity rate / 3.   - CPL = CPC / click-to-lead rate. Demand budget = leads x CPL.   - Capacity = opportunities needed / in-market accounts. Above 1, no budget closes the gap: the target needs more accounts or other channels. Show the share so the reader can judge it. 5. **Size the brand layer** (accounts not buying yet: the rest of the list).   - Impressions a month = audience x reach share x frequency. Brand budget = impressions / 1,000 x CPM.   - With a fixed budget, the brand layer gets what the demand layer leaves; show the frequency that buys at the reader's reach share.   - Budget above both layers' needs: don't raise frequency past the reader's target to spend it. Offer: a larger reach share (up to all of the audience), the next tier of roles, or returning it to `channel-fit`. 6. **Pick formats per layer** from the table in `references/linkedin-specs.md`. Brand: Thought Leader Ads, video, ungated documents. Demand: gated documents and lead gen forms with 1 qualifying question, retargeting of engaged accounts. Each format names the buying moment and the pillar it carries. 7. **Set the Thought Leader Ads slot.** Which voices (employees, experts, customers), the pick rules in the reference (organic baseline first, above the author's median, stands alone, claims checked, author agrees per post), and the share of the brand layer they get. For the posts themselves, `founder-to-ads`. 8. **Write the settings and exclusions** from the reference: audience expansion off, Audience Network off, cost caps, exclusions (customers, employees, competitors' employees, open opportunities from demand), the qualifying question, CRM stages through the Conversions API. 9. **Write the measurement.** Demand: CPL against max CPL in the first 2 weeks, cost per opportunity after the judgment window (1.5 x the sales cycle, from `lead-value-math`). Brand: reach and frequency inside the target accounts, engagement by role, and the pipeline from engaged accounts over the next 2 quarters, read against accounts the brand layer didn't reach. 10. **Run the sizing.** `python3 scripts/linkedin_budget.py` with the inputs (flags in its docstring). Without Python, do steps 3 to 5 by hand, show each formula with its figures, and say so. ## Check and verdict Check the plan against every line: | Check | Fails when || --- | --- || Every campaign has a layer, an audience, a format, a buying moment and a pillar | any is missing || Demand CPL at or under max CPL | above it (show by how much and the levers: offer, form, audience) || Capacity | opportunities needed exceed in-market accounts || Budget | layers cost more than the budget, or the brand layer gets nothing while most accounts are out of market || Figures | any audience size, CPM, CPC or rate without a source || Settings | expansion or Audience Network left on, no exclusions, no qualifying question || Copy in the plan | any line fails `voice-check` or `claim-check` (inline: no dashes, no banned words from the profile, every number in `## Claims and proof`) | | Verdict | Condition (first match wins) || --- | --- || `BLOCKED` | No target accounts or ideal customer that LinkedIn can target, and no budget or lead target to size from || `NOT YET` | Demand CPL above 2 x max CPL at the reader's own CPC and rate (the conversion rate would have to double, and no setting does that in a quarter), or capacity above 1 || `READY WITH GAPS` | The plan stands, but checks are still open after 2 rounds or inputs are `not available` (each named) || `READY` | Every check passes | Named stops: `BLOCKED` asks for the account list or the ideal customer. `NOT YET` stops and shows the gap, the levers in order (offer and form, audience, the ceiling's inputs) and the other channels to compare in `channel-fit`. No other stops. ## Loop 1. Run the checks above.2. Fix what failed: move budget between layers, change a format, add an exclusion, cut a figure without a source, rewrite a line.3. Check again. At most 2 rounds, then hand over with what is still open, labelled `open`. ## Learn Append to `## Learnings` when: - after the judgment window, the reader reports cost per opportunity by layer or format: a `fact`;- a role or format brings leads that never become opportunities: a `fact`, and a `proposal` to drop or change it;- the reader corrects the years between purchases: a `fact`;- the reader overrides the split: a `proposal` with their reason. Example: `- 2026-10-24 | linkedin-plan | fact | Gated document ads produced 0 opportunities in 2 quarters; demo forms produced 6 | Why: HubSpot, campaign source report, Q2 to Q3 2026 | Status: noted` ## Output ```Verdict: <READY | READY WITH GAPS | NOT YET | BLOCKED> · Coverage: <complete | partial (what's missing)> · Sizing: <script | by hand> Market: <n> target accounts · in market this quarter: <n> (<share>, from <years> between purchases)Need: <n> opportunities a quarter, <n> leads a month · capacity <share of in-market accounts> | Layer | Audience | Formats | Moment and pillar | Monthly budget | How it's judged || Brand | ... | ... | ... | ... | ... || Demand | ... | ... | ... | ... | ... || Total | | | | <sum> of <budget> | | Thought Leader Ads: <voices, rules, share of brand budget>Settings and exclusions: <list>Measurement: <first 2 weeks; after the judgment window>Check: <each check, pass or fail> · Rounds: <n>Open: <labelled items>Learnings applied: <list, or none>``` Then the handoff block. ## Hand off ```handoffskill: linkedin-planversion: 1.0.0verdict: READYcoverage: completefor: <company> LinkedIn Adsoutputs:  in_market_accounts: 100  demand_budget: 5775  brand_budget: 12960  formats: [thought leader ads, video, gated document, lead gen form]  layers: [brand, demand]  roles: [head of demand generation, vp marketing, revops manager]next: [ad-themes, founder-to-ads]open: []``` `ad-themes` reads `layers`, `roles` and `formats` to write themes per layer. `founder-to-ads` fills the Thought Leader Ads slot. ## Failure modes | Problem | Do this || --- | --- || The reader has no LinkedIn history, so no CPC or rate | Use Campaign Manager's forecast for the exact audience and label it `forecast`; never a benchmark || A fixed budget but no CPC, rate or forecast yet | Plan the first month as a learning read: the split is the reader's stated choice, labelled as such, and the layers are sized properly from the first month's own figures || The target list is under a few hundred accounts | Plan by account, not by audience size: name the roles per account and accept a higher frequency cap || The CRM can't send stages back | Judge demand on lead-to-opportunity from the CRM monthly and say the platform optimizes on form fills || The reader wants only lead gen | Show the out-of-market share and what happens to next quarter's in-market accounts without a brand layer; plan it if they still choose so, and record the choice || A Thought Leader Ad post carries a number without a source | Leave the post out until `claim-check` passes | ## Influenced by - The 95:5 rule from the B2B Institute, built on Ehrenberg-Bass research: most business buyers are out of market at any time, so brand reach and lead capture are planned as separate jobs. We size the in-market share from the reader's own purchase cycle rather than the headline figure: https://business.linkedin.com/marketing-solutions/b2b-institute/b2b-research/trends/95-5-rule- LinkedIn's Thought Leader Ads guidance, for who can be sponsored, approvals and recording the organic baseline first: https://business.linkedin.com/advertise/ads/sponsored-content/thought-leader-ads- SHAKEUP's LinkedIn budget method: lead budgets from cost per click and conversion rate, reach budgets from audience, reach, frequency and CPM, both held to the cost per lead ceiling.

How it thinks

Inputs in, a verdict out, and a stop only where you decide.

What we built

Most target accounts aren't buying this quarter, so we split the plan into reach for those and capture for the few that are. We size that split from your own purchase cycle and hold every lead to your cost per lead ceiling.

It reads your company profile if you have one, or asks for only what it needs. Then it works through its steps, checks its own work and ends in a verdict.

The steps

  1. 01

    Read your learnings first

    Formats or roles that produced opportunities before, or never did, change the defaults.

  2. 02

    Count who is buying now

    If accounts replace a tool like yours every 3 years, about 1 in 12 is in market each quarter. It uses your cycle, never a stock figure.

  3. 03

    Size the demand layer

    Your pipeline target becomes opportunities and leads, then a budget from your own cost per click and conversion rate.

  4. 04

    Check the capacity

    If you need more opportunities than there are accounts buying, no budget closes the gap, and it says so.

  5. 05

    Size the brand layer

    The accounts not buying yet get reach and frequency, priced from your CPM. Spare budget is offered back, not spent on extra frequency.

  6. 06

    Pick formats and Thought Leader Ads

    Each layer gets its formats, with sponsored posts picked against the author's own organic baseline.

  7. 07

    Check the plan

    Settings, exclusions, sources and the ceiling are each checked, fixed and checked again, at most twice.

What you need

  • Required

    Your target account list and buying group roles

    Your CRM → A CSV you upload → Your company profile

  • Required

    The most a lead may cost

    The handoff from lead value math → Your answer

  • Optional

    How often accounts buy or replace a tool like yours

    Your renewal and switching data → Your estimate, marked as one

  • Optional

    CPC, click-to-lead rate and CPM

    Your own past LinkedIn campaigns → The Campaign Manager forecast for the planned audience

  • Optional

    Audience size, reach and frequency for the brand layer

    The Campaign Manager forecast → Your answer

What it can return

  • READY

    Every check passes.

  • READY WITH GAPS

    The plan stands, with missing inputs or open checks named.

  • NOT YET

    Your own cost per lead on LinkedIn is over twice your ceiling, or you need more opportunities than there are accounts buying.

  • BLOCKED

    There's no account list or targetable buyer, and nothing to size from.

Where it stops for you

  • When there's no account list or buyer that LinkedIn can target
  • When the channel can't reach your ceiling or your capacity at your own figures

Example output

What you get back.

A run for a made-up company. Your output depends on your inputs.

Here's a plan for a made-up B2B SaaS company with 1,200 target accounts that switch tools about every 3 years. LinkedIn has to bring 30% of an $1.8M quarterly pipeline, and the most a lead may cost is $285.

Verdict

READY

Coverage: complete

Market

about 100 of the 1,200 accounts are in market this quarter (1 in 12). The plan needs 22.5 opportunities a quarter, so it has to win about 1 in 4 of the accounts that are buying.

LayerAudienceFormatsMonthly budgetWhat it's measured on
DemandBuying group roles at target accounts, engaged firstLead gen forms with 1 question, gated evaluation checklist$5,775 for 42 leads at $137.50Cost per lead now, cost per opportunity after 93 days
BrandAll 120,000 members in the roles, 60% reached 4 times a monthThought Leader Ads from the CEO and Head of Demand Gen, short product video$12,960Reach in target accounts, then pipeline from accounts it reached
Total$18,735 of $22,000
The other $3,265

held, not spent on extra frequency. Reaching 75% of the audience instead of 60% would use it.

Settings

audience expansion off, Audience Network off, cost caps, customers, employees and competitors excluded.

The handoff block the next skill reads
skill: linkedin-plan
version: 1.0.0
verdict: READY
coverage: complete
for: LinkedIn Ads, next quarter
outputs:
  in_market_accounts: 100
  demand_budget: 5775
  brand_budget: 12960
  formats: [thought leader ads, video, gated document, lead gen form]
  layers: [brand, demand]
  roles: [head of demand generation, vp marketing, revops manager]
next: [ad-themes, founder-to-ads]
open: [decide on the $3,265 not planned]

What good looks like

  • An in-market count built from your own purchase cycle
  • A demand budget that never plans leads above your ceiling
  • A brand layer for the accounts that aren't buying yet
  • No audience size, CPM or rate without a source
  • Audience expansion and Audience Network off, with exclusions set

Install

Use it in the tool you already have.

Pick your tool. Each tab has the shortest path we know of.

  1. 01
  2. 02
    In Claude, open Customize, then Skills.
  3. 03
    Choose +, then Upload a skill, and pick the ZIP. Code execution must be on in your settings.
  4. 04
    Ask for it by name, or let Claude use it when a task matches.

Questions

Before you add it.

Why split brand and demand at all?

Because most of your target accounts won't buy this quarter. Lead gen reaches the few that will. The brand layer keeps you on the shortlist of the rest for when they do.

Does it use the 5% in-market figure?

No. It uses the logic behind it with your own purchase cycle. A category bought every 3 years has about 8% of accounts in market each quarter, not 5%.

What if we've never run LinkedIn Ads?

It builds the plan and the formulas, and asks for the Campaign Manager forecast for your exact audience. Budgets stay not available until real figures exist.

Will it pick our Thought Leader Ads posts?

It sets the slot and the rules, like an organic baseline first and every claim checked. The founder posts to ads skill picks the posts.