How I Developed the CONTEXT Framework
I did not set out to create a framework. I set out to fix a problem that was costing my companies thousands of hours a year. The framework emerged from the solution.
By Tim Plummer, Creator of the CONTEXT Framework
The problem I kept seeing
I run businesses with AI every day. Not as a side project. Not as an experiment. As core infrastructure. My teams use AI for everything from market analysis to customer communications to operational planning.
And I kept seeing the same thing: 90% of professionals use AI like a search engine. One-shot, vague prompts. Disappointing results. Then they blame the tool.
“I tried ChatGPT but the output was generic.” I heard this dozens of times. Every time, the problem was not the AI. It was the prompt.
When I started training my own teams to use AI properly, I realised there was no structured system for writing prompts. Plenty of tips, plenty of “try adding more detail” advice, but no repeatable methodology that a professional could learn in an afternoon and use for the rest of their career.
So I decided to build one.
What I tried first
Before building anything new, I tested every existing framework I could find. I am not the kind of person who reinvents the wheel for the sake of it. If someone had already solved this, I wanted to use their solution.
I tested them all, systematically, across real business tasks:
Missed examples entirely. No mechanism for showing the AI what good output looks like. In practice, this is the single most impactful element.
Too focused on the creative writing use case. Missed nuance and constraints — the elements that matter most in business contexts where you need to say what NOT to do.
Too simple. Four elements is not enough to cover the dimensions that affect output quality. It works for basic prompts but breaks down on anything complex.
Better coverage, but awkward to remember and missing explicit format expectations. Teams struggled to recall the elements without a reference sheet.
Covered role, task, and format — but that is only three of the six dimensions that actually matter. No nuance, no examples, no tone control.
Each framework had genuine merit, but every one had gaps. And in professional settings, gaps mean inconsistency. I needed something comprehensive enough to handle any business prompt, but simple enough to remember without looking it up.
The six elements that actually matter
I did not sit down and design a framework on a whiteboard. The framework emerged from data.
Over several months, I analysed hundreds of real business prompts across my companies. I looked at what separated the prompts that produced excellent output first time from the ones that needed three or four rounds of revision.
Every time, the difference came down to the same six dimensions. Not five. Not seven. Six. When all six were present, the output was consistently good. When any one was missing, quality dropped in predictable ways.
“The framework was not invented. It was discovered. The six elements were already present in every great prompt I had ever written — I just had not named them yet.”
The acronym came last. Once I had the six dimensions, I needed a way to make them memorable. CONTEXT worked perfectly — not just as a mnemonic, but because the word itself captures the core principle: give the AI enough context and it will give you great output.
Why these six and not four or eight
Every element in CONTEXT exists because it solves a specific failure mode. Remove any one and output quality drops in a measurable, predictable way.
Without it, the AI makes assumptions about who you are and what situation you are in. Those assumptions are almost always wrong.
Without it, the AI guesses what you want. You get a vague essay when you needed a specific deliverable.
Without it, the AI oversimplifies. It ignores constraints, edge cases, and the things you specifically needed it to avoid.
Without it, the AI defaults to generic corporate voice. The same information in the wrong tone gets ignored by your audience.
Without it, the AI has no reference point for quality. Examples do what 100 words of instruction cannot — they show rather than tell.
Without it, the AI guesses at format, length, and structure. You get 500 words when you needed 150. You get paragraphs when you needed bullet points.
I tried versions with four elements. Quality dropped. I tried versions with eight. The extra elements added complexity without improving output. Six is the number that works.
From internal training to public framework
The first team I trained on CONTEXT was at Deus X Markets. The results were immediate and measurable.
Prompt success rate went from ~40% to ~85%
Measured as prompts that produced usable output on the first attempt, without requiring revision or re-prompting.
People who had been frustrated with AI were suddenly getting results. Not because the AI got better — because their prompts did. The framework gave them a checklist. Before hitting send, they could ask: “Have I covered all six elements?” If the answer was no, they knew exactly what was missing.
After training three teams internally, I realised this was not just a company tool. Every professional using AI was hitting the same wall. The framework needed to be available to everyone.
That is why I built Enigmatica.
The framework is free because AI literacy should be free
I could have packaged CONTEXT as a paid course. Plenty of people charge hundreds of pounds for less comprehensive prompt engineering training. But that would have defeated the purpose.
“AI literacy should not be locked behind a paywall. The professionals who need this most are often the ones with the least budget for training.”
Enigmatica is funded by enterprise training packages. Companies that want to train their entire teams on CONTEXT and other AI skills — with workshops, custom curriculum, and onboarding programmes — that is the business model.
The free platform is not a loss leader. It is the product. When professionals learn CONTEXT for free and see the results, they become advocates inside their organisations. That is how enterprise deals happen.
So yes, everything on Enigmatica is free. The CONTEXT Framework, all 40+ lessons, every tool. Register with your name and email and you have full access.
Because I believe the best way to sell AI training is to prove it works first.
Learn all six elements with examples and practice scenarios
One-page reference for the CONTEXT Framework
See the framework applied to real business prompts
How CONTEXT compares to CO-STAR, CRAFT, RACE, and others
Pass the 15-question assessment and earn your CONTEXT Certified badge
Build a CONTEXT-structured prompt step by step
Learn more about the founder of Enigmatica
Ready to Master the CONTEXT Framework?
The full CONTEXT Framework lesson is part of our Essentials level — with guided exercises, real-world practice, and a prompt grading tool. Completely free.