Account 360 Builder

Inside this Project, Claude carries the account picture as it develops:

• The org map as it is discovered, with each relationship marked confirmed or assumed, and upgraded as evidence arrives.
• Relationship history — who you have met, when, what was discussed, and where each person stands.
• The tech stack picture, including what is confirmed in use versus mentioned in passing.
• The account strategy as it evolves, including approaches already tried and how they landed.

One Project per account. This is the deliberate difference from Prospect Research Engine, which compares many accounts on one rubric; this one goes deep on a single account over time.

Reach for this Project when:

• An account is strategic enough to justify a standing workspace.
• Several stakeholders are involved and the org picture matters.
• You are preparing for a call and need everything known about the account in one place.
• You are handing an account over, or picking one up, and need the history to be legible.

Use something else when:

• You are scoring many accounts against an ICP — that is Prospect Research Engine.
• You are qualifying one opportunity inside the account — that is MEDDPICC Deal Strategy Workspace. Many accounts run both, one for the relationship and one per deal.

To start, you provide:

• The account name and why it is on your list.
• Research material — company information, news, filings, published strategy.
• Call notes from any conversations so far.
• CRM history if available: past opportunities, activity, prior owners.
• Stakeholder names and titles, including people you have not met.

Then you add after each interaction. An account workspace that is never updated is a document.

The integration reference matters more than it looks. Most tech-stack conversations turn on whether you connect to what they already run.

Claude accepts files up to 30MB each, and there is no fixed limit on how many you add, but everything in a Project's knowledge has to fit the context window. Fewer, sharper files beat a full archive.

Prepare files this way:

• Strip cover pages, legal boilerplate, and navigation chrome — they consume context and carry no signal.
• Prefer text formats. A clean export beats a scanned PDF.
• Name each file for what it contains, not when you exported it. Claude reads filenames.
• One subject per file. A merged reference document is harder to cite precisely.
• Date anything that goes stale, inside the file itself, so Claude can tell you when it is working from old material.

1. Create one Project for this account. Name it for the account — for example Account 360 — Ridgeline Group.
2. Paste the Project Instructions.
3. Replace the two bracketed fields. The tiering criteria determine how much strategy this account justifies; without them every account gets the same treatment.
4. Upload the knowledge files, persona library and integration reference first.
5. Run the first build with everything you have. Expect large gaps in the org map — that first gap list is your discovery agenda.
6. Update after every interaction with notes. The org map improves only if you feed it.

The bracketed fields below are placeholders you replace with your own details before you save the instructions. Leaving them unfilled produces generic output — that is the single most common reason a Project underperforms in its first week.

This Project has two:

• [FILL IN: Company Name] — your employer.
• [FILL IN: Your Account Tiering Criteria] — how your organisation classifies accounts, and what each tier warrants in coverage and investment. This is what lets Claude recommend a level of effort proportionate to the account rather than treating every account as strategic.

Role

You are a strategic account analyst for [FILL IN: Company Name], maintaining a living picture of one named account. Your value is that you never let a guess pass as a fact.

Operating rules

• Never infer reporting lines or influence from job title alone. Every edge is marked confirmed or assumed.
• Never record a relationship judgement as fact; cite the behaviour behind it.
• Distinguish technology confirmed in use from technology mentioned in passing.
• Never upgrade assumed to confirmed without new named evidence.
• Size strategy to the account's tier.

The seven-part 360

Company picture · business priorities in their own words · org map · tech stack · relationship history · account strategy · next actions.

The full instructions continue with the confirmed-versus-assumed rubric, the four commands, output formats, and escalation.

This Project produces four deliverables:

• Account 360 — the full seven-part picture.
• Org map — stakeholders, roles, stance, and influence, each marked confirmed or assumed.
• What changed — movement since the last update.
• Next actions — prioritised, each tied to a gap or an opportunity.

Org map and next actions. Ridgeline Group is fictional.

Org map — Ridgeline Group · updated 9 August 2026

Two of six relationships are confirmed on influence; four are assumed. Both high-influence assumed nodes — security and finance — are unmet, which is the shape of the risk here.

Next actions, prioritised

1. Get to Okafor. A security lead who blocked a prior vendor is the most consequential unknown in the map. Ask Lindqvist for an introduction framed as an early technical review rather than a sales conversation — early security engagement de-risks the timeline.
2. Confirm Whitfield's approval threshold. Ask Achebe directly what amount requires CFO sign-off. Cheap question, removes an assumption on a high-influence node.
3. Address Park's migration concern. Low influence, but a repeated objection spreads. Offer the migration reference from the knowledge files.
4. Clarify Bianchi's role before procurement becomes urgent rather than after.

Assumed influence is not a weaker version of confirmed. It means we do not know.