How Threader works
Strategy is a commitment, not a conversation. Threader is built to enforce that distinction. This page explains the thinking behind the four products, the methodology embedded in the tools, and how the technology is built.
Tools that decide, not tools that converse
Most AI tools optimise for fluency. They produce confident-sounding output across any prompt, in any direction, with no resistance. The result is what HBR researchers Romasanta, Thomas and Levina recently called trendslop: synthesised consensus that sounds strategic and decides nothing.
Strategy is the opposite of that. It is the act of choosing what not to do. It demands a diagnosis, a guiding policy, and a coherent set of actions, in that order. It requires trade-offs that close off options. It rewards conviction and punishes hedging.
Threader is built around that distinction. The four products share a single product-level promise: decision architecture for the discipline you work in. Brand strategy. Communications planning. Employer brand. What comes next in your career. Each product is structured so the user cannot skip the diagnosis, cannot avoid the trade-off, cannot generate the next layer until the previous one is locked.
Built on established frameworks, not invented yesterday
The frameworks embedded in the tools are not new. They are the established frameworks of evidence-based strategy in marketing, communications, employer brand, and career development. The same frameworks currently championed by the IPA, WARC, Marketing Week, the CIPR, the CIPD, the Ehrenberg-Bass Institute, and the leading academic voices in career construction theory.
The cascade across the three B2B products
The sequence the tools enforce is Mark Ritson’s segmentation, targeting, positioning discipline. You cannot target before you have diagnosed. You cannot position before you have targeted. You cannot generate creative territories before you have positioned.
The diagnosis layer applies Richard Rumelt’s kernel from Good Strategy Bad Strategy and The Crux. Every cascade starts with a reframe of the brief. The exclusion discipline at the positioning step applies Michael Porter’s principle that strategy is choosing what not to do; every cascade requires an explicit exclusion before it will continue.
The creative layer applies Byron Sharp’s mental availability theory and Jenni Romaniuk’s operational work on distinctive assets, both from the Ehrenberg-Bass Institute. The whole cascade reflects Les Binet and Peter Field’s effectiveness research for the IPA Databank. The audience layer applies Alex Murrell’s tension-first approach. The convention-breaking layer pairs Jean-Marie Dru’s disruption methodology with Adam Morgan’s challenger thinking.
Where each product extends the frameworks
The governing principle
These foundations are embedded in the prompts and the cascade structure. They never appear in the client-facing output. Users feel the discipline; they do not read a theory lesson. The tools are opinionated scaffolding for the strategist’s judgement, not a textbook.
Grounded in this week’s reality, not last year’s training data
Every strategic output Threader produces is grounded in current context. A weekly macro refresh updates the underlying market conditions every product reasons against. Per-category live validation runs on every cascade, so the work is checked against the state of the category at the moment the user is working in it.
The macro context layer is a public artefact. You can see it at macro-context.json. This is not a marketing claim about being current. It is the file the system reads, exposed to anyone who wants to verify it.
The distinction matters. A generic LLM reasons from training data that may be many months old. Threader reasons from a context layer refreshed weekly and validated at the point of use.
Disciplined engineering, made visible
Threader is built and maintained by a small UK-based team. The technology is AI-assisted, human-led: every line of code is reviewed, tested, and understood. Changes follow a structured release process. The architecture is deliberate.
Each product runs on isolated worker infrastructure, so a fault in one tool cannot cascade into another. Data is structured, versioned on save, and exportable at any time. The platform status page is public and the per-product changelogs are open: anyone can see what has shipped, when, and what was fixed.
What Threader will not do is just as important. It does not replace strategists. It does not produce finished creative work. It does not make strategic decisions for you. It structures the thinking, enforces the trade-offs the frameworks advocate, and produces outputs grounded in the research the field publishes. The judgement is yours.