agentleFS
CSPulseCSPulse/customer-success-skills/skills/account-plan/SKILL.md

account-plan

Skill2 starsChanged 3 months agoMITView on GitHub

Use when the user says "account plan", "strategic account plan", "twelve month plan", "what is our plan for [account]", "plan for this account", "write up the account strategy". Produces a one to two page plan a successor could run in month four: the objective in the customer's words with a baseline and target, owners on both sides, 3 to 5 dated actions and risks with triggers. Not for the first 90 days of a new customer (use onboarding-plan), a renewal call (use renewal-risk) or choosing which accounts earn a plan (use book-triage).

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the user gave the account, the contract (ARR, renewal, notice period) and what the customer is trying to achieve, go straight to "Do …

    line 13
  2. Very specific:

    Use this skeleton. Keep it to two pages.

    line 35

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Account Plan

A durable plan for one account that someone inheriting it in month four could read and know what was promised, by whom, and what should be true by when.

Start here

If the user gave the account, the contract (ARR, renewal, notice period) and what the customer is trying to achieve, go straight to "Do this".

Otherwise ask for these in ONE message, then proceed with whatever comes back:

  1. Account, ARR, renewal date, notice period
  2. What the customer says they are trying to achieve, in their words, and who said it
  3. Who decides, who champions, and who has not been met

Anything else (usage, the original business case, emails) is a bonus. If an "Account context" block exists in the conversation, use it for the value metric and what healthy looks like here.

Do this

  1. Write one or two objectives in the customer's words. "Cut dispute resolution from 14 days to 5" is theirs. "Drive adoption" is yours. Give each a baseline with the date it was taken, a target with a date, and a named owner on their side.
  2. If no measure exists, say so first. Agreeing the measure becomes action one. An account with no agreed measure is the common shape of a healthy-looking renewal that fails.
  3. State where it stands in four lines: what works, what does not, what changed since the last version, what is not known. Tag facts said, seen or inferred.
  4. Summarize the people: decider, champion, sceptic, the biggest gap and the route to close it, and who is on the account from our side. Point to a stakeholder map if installed rather than rebuilding it.
  5. Write 3 to 5 actions. Each has an owner, a date, the objective it serves, and at least two belong to the customer. Recurring items (QBR cadence, monthly report) go under Rhythms, not here.
  6. Give every risk a trigger. What would tell you it is happening, what you will do, by when you need to know. Include single-threading and no agreed measure if they apply.
  7. Set review triggers, not just a cadence: sponsor change, contraction, escalation, missed measure, their reorg. Each review answers what moved, what did not and why, and what changes in the plan.

Output

Use this skeleton. Keep it to two pages. A fill-in version is assets/account-plan.md.

ACCOUNT PLAN: [Account] | [ARR] | renews [date] | opt-out by [date] | written [date]

Plan in one line: [objective, measure, biggest risk]

Objective: [in their words] | Measure: [baseline, date] to [target, date]
Their owner: [name, role]. Agrees it is theirs: yes / no / not asked

Where it stands
- Working / Not working / Changed / Not known (each tagged said, seen, inferred)

People: Decider [name, met?] | Champion [..] | Sceptic [..] | Gap and route [..]

Actions: [action | owner | side | date | objective served]
Rhythms: [recurring items]

Risks: [risk | trigger | response | know by]

Review: [cadence] and forced updates on [events]

Commitments on record: [what | by whom | side | date said]

What I could not check
- [gap]

A filled example is in examples/sample.md.

Rules

  • A plan with no measure cannot be failed, so it is never achieved. Activity lists feel full and prove nothing at renewal.
  • A measure without a baseline cannot show movement. At the renewal you would argue about whether anything improved.
  • An objective nobody on their side owns is yours. It will not survive a change of sponsor.
  • Five actions or fewer. More than that and nothing is prioritized, and the plan stops being opened.
  • Actions that are all ours mean the customer never agreed. Some must be theirs, agreed with them.
  • A risk without a trigger is a worry. Triggers are what let you notice it early enough to act.
  • Date everything and attribute commitments. A commitment with no name is unenforceable both ways, and month-four you cannot tell what was told from what was concluded.
  • Not every account earns a plan. A plan for all forty accounts is how the format loses credibility.

Related (only if installed)

book-triage to decide which accounts earn a plan. stakeholder-map for the people picture. renewal-risk for whether the plan is working. business-review for the meeting where it is agreed with the customer.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.