Everything you need to get started with Claude at Wilbury Stratton. Work through each module in order and refer back whenever you need a refresher.
How to access Claude, choose the right plan, and understand the basic interface — everything you need before your first conversation.
Type your message in the box at the bottom and press Enter (or click the arrow). Claude responds in the main area. Each conversation is self-contained — Claude remembers everything within one chat window.
Your conversation history lives in the left sidebar. Each chat is saved automatically. You'll also find your Projects here — more on those in Module 06.
Click the pencil/compose icon (or "New Chat") to start a fresh conversation. Each new chat starts with a blank slate — Claude won't carry anything over from previous chats unless you're using memory features or Projects.
Click your profile icon (top right or bottom left depending on your version) to access Settings. This is where you'll set your Personal Preferences, manage Memory, and adjust your account — all covered in the next modules.
Set your instructions once — Claude follows them automatically in every conversation. Think of it like briefing a new colleague on how you like to work.
Personal Preferences are instructions you give Claude once, and it remembers them across all conversations. Think of it like briefing a new colleague on how you like to work — you only need to do it once, and from that point on they adapt to you automatically.
Click your profile icon — top right on desktop, or bottom left on some versions — and select Settings.
Look for a section labelled Profile or User Preferences. This is where you write instructions that Claude follows automatically in every conversation.
Type your preferences in plain English — be specific. For example: "I prefer bullet points over long paragraphs. I work in recruitment. Keep responses concise and use British English."
Click Save, then start a new conversation and notice how Claude automatically adapts. You can update your preferences at any time — changes apply to all new conversations going forward.
"Sure! Here's a comprehensive overview with many important considerations to keep in mind. First, it's important to note that..." [generic, long, not tailored to you or your industry]
Recommendation: [Direct, tailored answer]
One of the most powerful features of Preferences is the ability to create custom trigger phrases. You teach Claude that when you type a specific word, it should always respond in a specific predetermined way — like keyboard shortcuts, but in plain English.
Learn how Claude remembers — and when it doesn't. Understanding this prevents frustration and helps you get consistent results every time.
Claude has two types of memory: within a conversation (it remembers everything you've said in that chat window) and across conversations (this only persists if memory features or Projects are enabled). Understanding this prevents frustration and helps you plan your workflows effectively.
Personal Preferences, Memory feature saves, Project context
Everything said in one open conversation window stays active throughout
New conversations start completely fresh unless memory features are switched on
Inside a single chat, Claude remembers everything. You can refer back to earlier messages, build on previous answers, and have a flowing, natural conversation. The longer the chat, the more context Claude has.
When you open a new chat, Claude starts with a blank slate. It won't remember your previous conversations unless memory features are switched on or you're working within a Project.
Claude has a memory system that can store key facts about you across separate conversations — your job title, working preferences, and ongoing projects. View, edit, or delete these in Settings → Memory.
Claude's Projects feature creates a dedicated workspace where context is preserved across all conversations within that project. Ideal for ongoing client mandates and team workflows needing consistent context.
At the start of any important new conversation, briefly remind Claude of relevant context: "I'm working on a proposal for a senior finance client. Here's their brief..." — this takes ten seconds and ensures Claude always has exactly what it needs.
Get Claude working like a well-briefed member of the WS team — tailored to your industry, your clients, and how you communicate.
Without context, Claude gives generic answers. With business context, it delivers responses tailored to your specific work — whether that's structuring a talent landscape briefing, drafting a candidate outreach message, or producing a market intelligence report. Think of the difference between a well-intentioned colleague who's never worked in executive search, versus one who genuinely understands what WS does and how we do it.
The simplest approach: add your business context to your Personal Preferences in Settings. Claude will then automatically know who you work for and what you do in every conversation.
Use Claude's Projects feature to create a dedicated "Wilbury Stratton" project with a system prompt covering WS's services, your role, tone of voice, and typical outputs — always ready when you need it.
Cover these key areas for the richest, most useful context:
For specific tasks, add a quick context note at the top of your prompt: "Context: I'm working on a talent intelligence report for a FTSE 100 client in financial services. The mandate is a Group CRO search. Tone should be consultative and senior."
"I work at Wilbury Stratton, a global talent research and intelligence firm with offices in Brighton, London, New York, Kuala Lumpur, and Singapore. My role is [your role] on the executive search team. We work on a retained basis, placing Director to C-Suite talent across all sectors globally. My typical tasks include: structuring mandate briefs from client calls, drafting candidate outreach and engagement messages, producing shortlist reports, and preparing candidate interview briefs.
Our tone is professional, consultative, and direct — we position ourselves as trusted advisers, not transactional recruiters. Always use British English. When writing externally, assume the reader is a senior executive who is time-poor and not actively looking."
"I work at Wilbury Stratton, a global talent research and intelligence firm. My role is [your role] on the research and intelligence team. Our work includes talent mapping and pipelining, market and competitor intelligence, strategic intelligence, perception analysis, diversity benchmarking, and succession planning. We serve over a third of the FTSE 100 and work globally across all sectors.
My typical outputs are talent landscape reports, market intelligence briefings, org mapping documents, and sector analysis. Outputs should be structured, evidence-based, and written for a senior audience. Always use British English. Flag clearly when anything in a response should be verified before sharing with a client."
Master the GCAO framework — Goal, Context, Action, Output + Iterate. The difference between a weak and a strong prompt is the difference between a generic draft and a publish-ready result.
A prompt is the message or instruction you give Claude. Better prompts produce better results — every single time. Think of prompting like briefing a brilliant freelancer. The clearer your brief, the better the output. Vague in, vague out. Specific in, brilliant out.
Start with the end in mind. A clear goal frames everything that follows.
Give Claude the background: who is involved, what the situation is, any relevant constraints or history. More context = more tailored output.
Be specific about the task. Use clear action verbs: write, summarise, analyse, compare, suggest, draft, rewrite, list, critique, simplify.
Specify the format, length, tone, and style you want. This prevents Claude from guessing and producing something that needs heavy editing.
Treat Claude's first response as a solid draft, not a final answer. Ask it to adjust tone, shorten, be more specific, or try a different angle. Each iteration gets you closer to perfect — and it's fast.
I want to turn my notes from a client briefing call into a structured mandate summary, then use it to write a compelling outreach message for target candidates.
I've just spoken with the hiring manager at [client name]. My notes are rough — stream of consciousness from the call. The role is a [title] and the key things I picked up are: [paste your rough notes here]. The ideal candidate is currently employed, probably not looking, and will need a compelling reason to engage.
First, extract the key points from my notes and organise them into a clean mandate summary with sections: Role Overview, Why This Opportunity, Ideal Candidate Profile, Key Challenges, and Compensation. Then, using that summary, write a short outreach message I can send to a target candidate on LinkedIn.
Mandate summary: structured, bullet points per section, under 300 words. Outreach message: under 120 words, warm and direct, focus on the opportunity not the job spec, British English, end with a soft call to action. Do not make up any details — if something is missing from my notes, flag it with [MISSING: X].
"The outreach message is good but too formal. Make it sound more like a message from a trusted advisor than a recruiter. The second paragraph is too long — tighten it to two sentences."
I want to build a talent landscape briefing on the Chief Risk Officer market within UK retail banking for a client pitch.
Our client is a mid-sized challenger bank planning to hire a CRO for the first time. They have no benchmark data on compensation, tenure, or where CRO talent typically moves from or to. The pitch is in 5 days and needs to demonstrate WS's market knowledge.
Analyse the CRO talent landscape in UK retail banking. Identify typical career paths into and out of CRO roles, likely candidate profiles, key feeder organisations, and any market trends worth flagging.
Structured briefing document. Sections: Market Overview, Typical Candidate Profile, Feeder Organisations, Compensation Benchmarks (ranges), Key Trends. Bullet points throughout. Professional, consultative tone. Flag anything I should verify before sharing with the client.
"Good — now add a section on 3 specific individuals who are publicly known to have recently moved into or out of CRO roles at UK retail banks. For each, note their previous employer, current role, and why they might be relevant to mention in a pitch."
"Write an outreach message for a finance candidate."
No mandate context. No candidate profile. No tone guidance. Result: a generic template that sounds like every other recruiter.
[Goal] Turn call notes into a mandate summary + outreach message.
[Context] Paste rough notes from client call.
[Action] Extract key points, then write a LinkedIn message.
[Output] Bullet summary + 120-word message, flag anything missing.
"Tell me about the CFO market."
Too vague. Claude will give a textbook answer, not an intelligence briefing.
[Goal] Build a CFO talent landscape for a fintech client pitch.
[Context] Series B company, 80 people, London.
[Action] Identify candidate profiles and feeder orgs.
[Output] Structured briefing, bullets, flag anything to verify.
Projects are Claude's most powerful feature for professional work. Set up once, use consistently — for every mandate, every client, every team member.
A Project is a dedicated workspace inside Claude that keeps all your instructions, context, and uploaded files in one place — and applies them automatically to every conversation within that project. Think of it as a permanent, pre-briefed assistant that already knows the mandate, the client, and the way you want to work before you've typed a single word.
Instructions and documents stay active across every conversation in the project
Attach briefing documents, CVs, reports — Claude reads them automatically
Share a project with colleagues so everyone works from the same brief
In the Claude sidebar, click New Project. Give it a clear name — e.g. "Barclays — Group CRO Search" or "HSBC Talent Mapping Q3". Use the client name and mandate type so it's instantly recognisable.
Click Set project instructions. This is your system prompt — the standing brief Claude reads before every single conversation in this project. Be specific and comprehensive.
Add any reference files the project needs — the client brief, mandate specification, org charts, previous reports, sector context. Claude will reference these automatically across every conversation.
Click the Share option within the project to invite colleagues. Set permissions so others can use the project, or also edit instructions and files.
Every new conversation opened inside the project automatically inherits all your instructions and files. No re-briefing, no copy-pasting context — just start asking.
The most effective way to use Projects at WS is to create one dedicated project for every active mandate. All Claude work related to that mandate — research, outreach, profiles, report writing — is done inside a single, fully briefed workspace.
"You are supporting a Wilbury Stratton executive search project. Here are the key details:
Client: [Client name] — [brief description of the organisation]
Mandate: [Role title] search — [brief description of the role and its purpose]
Ideal candidate: [Key criteria — background, experience, competencies]
Sector: [Industry/function]
Geography: [Location scope]
Sensitivities: [Anything to be mindful of — off-limits companies, confidentiality requirements, etc.]
WS tone: professional, consultative, and direct. Always use British English. When writing candidate outreach, assume the reader is a senior executive who is not actively looking. Flag anything you are uncertain about rather than guessing."
Inside any project, click the share icon and add colleagues by email. Everyone works from the same instructions and files — no briefing the team individually, no version control issues.
Can use — they can chat within the project but not change instructions or files. Can edit — full control. For most mandates, give the research team "Can use" and keep instruction editing with the project owner.
As a mandate evolves — new criteria, updated client feedback, additional sensitivities — update the project instructions so the whole team automatically works from the latest brief. One update, everyone benefits instantly.
One project per search. Include the full brief, ideal candidate profile, client background, off-limits companies, and tone guidance.
Upload the target organisation list, mapping criteria, and sector context. Use for longlist building, profile drafting, and insight synthesis.
Include the pipeline criteria and client brief. Track candidate engagement notes and keep CVs uploaded as they come in.
Upload all research materials, the client brief, and any previous deliverables. Use for drafting sections, synthesising findings, and writing the executive summary.
A standing project for a major repeat client. Include their preferences, communication style, org context, and history of past work — always ready when they call.
A team-wide project for internal tasks — proposals, templates, standard language. Everyone accesses the same resources.
"Using one generic Claude project for all mandates" — Claude can't distinguish between clients and context becomes muddled.
"Never updating project instructions" — stale briefs mean stale outputs, especially as mandates evolve.
"Uploading every document available" — too many unrelated files dilutes the quality of responses.
One project per active mandate — clean, focused, and fully briefed from day one.
Update instructions when the brief changes — one edit updates the brief for the whole team instantly.
Upload only what's relevant — the client brief, mandate spec, and key reference documents. Keep it tight.
Quick-reference definitions for key Claude and AI terms — plus WS-specific industry terminology.