AI Skills Docs
Reference

Context: The Knowledge Base for AI

Context is the difference between AI that gives generic answers and AI that tailors answers to your specific situation.

Core Skill

Context management

AI has no memory and doesn't know your situation. Every chat starts with a blank slate. Those who learn to provide the right context get output that directly aligns with their organization, team, and specific task. Context management is the difference between generic answers and usable results.

Colleague vs. Language Model

To understand why context is so important, consider the difference between a human colleague and an AI language model:

Colleague
Language Model
Knows your organization No memory between sessions
Understands your jargon No knowledge of your situation
Familiar with the history Guesses at meaning
Gets implicit expectations Uses general patterns

How AI models work with information

AI models are trained on large amounts of text and work based on two sources:

What AI learned during training

The model is trained on billions of words of text up to a certain date (knowledge cutoff). This forms its general knowledge of language, facts, and patterns.

What you explicitly provide

All information you add in your prompt and via attachments (files, documents). This is the only way AI knows your specific situation.

Important to remember

Recent information after the training date is invisible to the model. AI doesn't know your specific organization, colleagues, internal processes, team structure, or current projects: unless you explicitly provide this as context.

The Context Formula

The more specific the input, the more relevant the output. The formula is simple:

Task

goal, who, what, why

+

Background

organization, team, project

=

Optimal Result

In the Context Formula, task context is always your starting point. Then add background about your specific situation. The more targeted your background, the more usable the result.

The difference context makes

Without Context

Prompt

Write a proposal for remote work at our healthcare organization

Result

  • Generic American proposal
  • Standard terms: "Remote work"
  • General office rules
  • Outdated regulations
With Context

Prompt

Write a proposal for hybrid work for our healthcare organization with 1200 employees, healthcare collective agreement, and 24/7 services across 3 locations.

  • Expertise: HR advisor
  • Audience: Works council and management
  • Focus: Patient safety
  • Documents: @Organization.docx

Result

  • Specific for healthcare sector
  • Collective agreement conditions
  • 24/7 continuity safeguards
  • Works council language and reasoning

Expert Advice

1

Create a context document

Write a document with your organization, team, your role, and key terminology. Save this for reuse via attachments.

2

Referring to a website?

AI doesn't always read the right info from websites. Copy the relevant text and add it directly to your prompt.

3

AI has limited memory

Too much context can lead to poor results. Only provide information that's relevant to the specific task.

Context in practice

For professionals, here are concrete examples of valuable context:

Team & organization context

Team description, brand guidelines, tone of voice per brand, stakeholder list, approval processes.

Project context

Campaign goals and KPIs, target audience personas, budget and timelines, previous results, competitive analysis.

Role-specific context

For marketers: media mix preferences, channel strategy. For content managers: SEO guidelines, content calendar. For CRM: segmentation criteria, flow templates.

Managing context

Just like in a human conversation: the longer the conversation, the harder it becomes to remember everything. AI models have a context limit (often 100k-200k tokens, or ~75,000-150,000 words).

Start with a clean chat

Start new tasks in a new chat. Old context from previous conversations can unintentionally influence your new output.

Be selective with attachments

Only add relevant documents that directly help with your specific task. Not your entire archive: too much context reduces quality.

Split complex tasks

Create multiple focused prompts instead of one long one. This keeps the context clear and improves results per subtask.

Save successful prompts

Copy working prompt+context combinations to a prompt library for reuse.

Want to learn more about context?

In the AI Fundamentals Training, you'll learn how to set up a structured AI work environment with reusable context documents. View the training →

Summary

  • 1. AI has no memory: Every chat starts blank. Always provide the context AI needs.
  • 2. Context = quality: The more specific your input, the more relevant the output.
  • 3. Build a context library: Create reusable context documents for your organization, team, and projects.
  • 4. Be selective: Too much context is just as problematic as too little. Focus on what's relevant.