Personal Productivity Sales & Outreach Knowledge Management

Hiten Shah's AI use case

Serial founder at Nira / Crazy Egg / KISSmetrics

Hiten Shah uses ChatGPT Projects as context-loaded assistants for managing up, self-coaching, and turning sales frameworks into custom discovery scripts.

The problem

What was broken before AI

Managing up, coaching yourself through interpersonal situations, and applying dense sales frameworks all depend on context that is easy to forget, misread, or leave scattered across documents, articles, and past conversations.

What changed

What the use case made possible

ChatGPT Projects let him keep relevant files and instructions inside persistent workspaces, so each assistant can reuse the same source material across conversations.

Why this matters

Why this use case is worth studying

Most people try to get better AI answers by polishing the final prompt. Hiten’s approach moves the leverage upstream: gather stronger reference material, turn it into project instructions, then use the assistant in real work situations where context matters. That makes AI less like a one-off answer machine and more like a reusable operating layer for judgment-heavy work.

Use this when

When this pattern applies

Use this when you repeatedly ask AI for advice or deliverables that depend on background context: a manager’s working style, your own communication patterns, a company sales framework, or product-specific knowledge.

Exponential Builder analysis

Context is the real interface.

Hiten’s setup shows that better AI output often comes from better inputs before the chat starts: operating manuals, methodology docs, voice guides, and product context shape the assistant more than a clever last-mile prompt.

Reuse turns judgment work into a system.

Managing up, self-coaching, and sales discovery all require recurring interpretation of messy human context; putting that context into separate Projects makes the assistant easier to trust, test, and improve over time.

Narrow assistants compound faster.

A manager-prep Project, a Personal OS, and a sales discovery Project each have a clear job, which makes it obvious when the assistant is missing source material or drifting into generic advice.

Who this is for

Best fit

Founders who want better prep for high-stakes conversations

Operators and PMs who manage up often

Sales leaders turning methodology docs into usable scripts

Managers building personal coaching or reflection workflows

Knowledge workers who keep re-explaining the same context to ChatGPT

What to avoid

Mistakes and warnings

Where this pattern can go wrong if you copy it too literally.

Do not upload confidential operating manuals, private personality assessments, internal sales materials, or customer data unless you have permission and the right account/privacy setup.

Be careful with “AI clone” language; treat the assistant as a preparation aid, not a replacement for the real person’s judgment.

Personality frameworks can be useful prompts for reflection, but they are imperfect and should not be treated as clinical or scientific assessments.

A sales-methodology assistant can generate plausible scripts that still miss the actual buyer, market, or legal constraints.

If the assistant sounds generic, the fix is often better source material rather than a more elaborate prompt.

Review anything customer-facing before sending it.

Public workflow preview

The shape of the workflow

A high-level look at how the use case works, with the reusable pattern made clear.

Pick one recurring decision

Choose a situation where context changes the quality of the answer, such as pitching your manager, handling conflict, or running sales discovery.

Gather the source material

Add operating manuals, shared articles, personality notes, sales methodology PDFs, or product context.

Generate the project instructions

Ask ChatGPT to turn the uploaded material into instructions for the assistant’s role, boundaries, and response style.

Test with a real scenario

Use an actual upcoming conversation, discovery call, or interpersonal challenge instead of a toy example.

Refine with examples

Add better inputs when the answer feels vague, off-tone, or too generic.

Copy the pattern

The reusable idea

Pattern in one sentence

Create narrow ChatGPT Projects, load them with high-quality context, then let each workspace become a reusable assistant for one recurring class of work.

Reusable idea

Start with one assistant that has a narrow job. A manager-prep project, a personal coaching project, or a sales-methodology project will teach you more than a giant catch-all workspace. The practical trick is to give the assistant source material that represents the standard you want it to follow: how someone communicates, how a framework works, what your product does, and what a good answer should feel like.

Steal this workflow

Create one dedicated ChatGPT Project for a recurring high-context task.

Checklist

Pick one narrow use case: manager prep, self-coaching, or sales discovery.

Upload 3–10 reference materials that represent the standard you want followed: operating manuals, articles, voice notes, sales methodology PDFs, product context, ICP notes, or examples of strong outputs.

Ask ChatGPT to inspect the files and draft Project instructions covering the assistant’s role, what sources to rely on, response style, boundaries, and what to ask when context is missing.

Paste the instructions into the Project and edit for privacy, clarity, and scope.

Test it on a real upcoming situation, such as pitching an idea, preparing for a difficult conversation, or building a discovery guide.

When the answer feels vague, add better source material or a concrete example before rewriting the prompt.

Keep separate Projects for separate jobs so the context stays clean.

Suggested prompt

“I’m creating a dedicated ChatGPT Project for [manager prep / self-coaching / sales discovery]. Use the uploaded files to draft clear Project instructions for an assistant that will help with [specific recurring task]. The instructions should define the assistant’s role, which source materials to rely on, the response style to use, what boundaries to follow, and what clarifying questions to ask when context is missing. Base the instructions only on the provided material, and make them practical enough to paste into the Project settings.”

Field notes

Get new AI use cases in your inbox

A short weekly note on how real people are using AI to save time, make money, build tools, and run their lives.

Related use cases

Keep exploring nearby systems.

Browse all

Customer SupportSales & Outreach+1

Minerva CQ team

AI Research Division / product team at Minerva CQ

Real-time contact-center co-pilot workflow where AI listens to live voice support, detects intent and sentiment, retrieves context, updates customer profiles, and guides agents through next-best actions.

Minerva CQOpenSearchLLM APIs

Use case

View use case

Business OperationsKnowledge Management+1

Included Health clinical team

Clinical team using an internal ambient scribe workflow at Included Health

Telehealth ambient scribe workflow where visit audio is transcribed with Whisper and processed with GPT-4o into SOAP notes and patient instructions for clinician review.

WhisperChatGPTOpenAI Audio API+5

Use case

View use case

Sales & OutreachBusiness Operations+1

Oracle AI Sales Agents

AI agents for sales administration and account intelligence at Oracle

Sales operations workflow where AI agents help sales professionals update company records and generate reports from varied data sources, including multilingual information.

Oracle Fusion Cloud SalesOracle Fusion Cloud Customer Experience (CX)

Use case

View use case