Context Engineering

This data was originally featured in the July 9th, 2025 newsletter found here: INBOX INSIGHTS, July 9, 2025: Is AI Use Cheating, Context Engineering

In this week’s Data Diaries, let’s break down a new AI term that you’ve probably seen the nerds talking about if you hang out in the various AI-forward chats: context engineering.

What, you might ask, is this fresh new hell? What does it mean, and is it something you should care about?

Context engineering states that what separates great AI results from mediocre ones is not prompting, but context. Coined by OpenAI cofounder Andrej Karpathy, context engineering is… not new. In fact, it’s something we’ve taught in our Generative AI for Marketers courses for over a year now, just under a different name: knowledge blocks.

What is a knowledge block? In prompt engineering, knowledge blocks are exactly what they sound like – blocks of information that you have on hand when you’re using your favorite AI tools.

Why? Because – and this has been true since the earliest days of generative AI, since GPT-2 back in 2020 – the more relevant, specific words you use in a prompt, the better AI performs. The very nature of the tools dictates this; at their hearts, today’s language models are prediction engines. The more relevant information they have to predict with, the better their predictions tend to be.

The easiest way to provide a lot of relevant words? To build them in advance, like Legos, and then add them to prompts as you need them, from ideal customer profiles to how you do your marketing to SWOT analysis of your competitive space.

And today’s AI models can hold enormous amounts of information in a single prompt. ChatGPT can accommodate an entire business book of 50,000 words in a prompt easily. Google’s Gemini can handle the collected works of William Shakespeare, all 800,000 words, in a prompt.

The key takeaway here is that you should be building these knowledge blocks, storing them in an accessible place for you and your team, and dropping them into prompts any time you’re doing work where precision and correctness are important. The more data you bring to the party, the better AI tends to perform.

And if you’d like to learn how to build knowledge blocks that are robust and powerful, we teach that in our Generative AI Use Cases For Marketers course.

Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.