Why I Ended Up Shaping My Procurement Persona While Building The HITSS Solution.

One of the recurring assumptions in AI discussions is that access to a large language model automatically creates expertise. It does not.

A large language model has access to a vast amount of information, but information and expertise are not the same thing. Expertise is built through experience, repeated decision-making, lessons learned, successes, failures, and an understanding of the context in which decisions are made.

Procurement professionals understand this intuitively. Two category managers can read the same supplier proposal and arrive at different conclusions. Two sourcing specialists can evaluate the same market conditions and develop completely different strategies. The difference is rarely access to information. The difference is experience, judgment, and commercial reasoning.

This observation became one of the added benefits of creating the HITSS (Hybrid IT Sourcing Strategy) Solution.

The Real Problem Is Not Lack of Knowledge

Over the years, you have created sourcing strategies, category plans, negotiation records, supplier performance assessments, governance frameworks, contract templates, business cases, training materials, presentations, lessons-learned documents, and operational procedures. You may also run a professional blog as I do.

The problem is that all that knowledge is fragmented.

• Some of it exists in SharePoint.

• Some of it sits in old PowerPoint presentations.

• Some of it is hidden in contract files.

• Some of it exists only in Teams conversations or email threads.

• A significant portion exists only in your head.

The result is a situation I have seen repeatedly throughout my procurement career.

We continuously recreate knowledge we already possess.

The same supplier issues are rediscovered. The same governance discussions are repeated. The same commercial mistakes reappear every few years because the people involved in the original decisions have moved on.

What we call “knowledge management” is frequently little more than document storage.

Knowledge storage and knowledge utilization are not the same thing. (c) Captain Obvious.

Why Procurement Is Vulnerable

This challenge is especially visible in procurement. Unlike many operational functions, procurement decisions are heavily influenced by context. A sourcing strategy that was appropriate three years ago may be completely inappropriate today.

Much of this context never appears in structured systems. It exists in discussions, presentations, emails, minutes, and your accumulated professional experience.

This creates an interesting paradox. Procurement functions often have enormous amounts of information available to them, yet still struggle to apply organisational knowledge consistently.

The issue is not the absence of information. The issue is the inability to connect information with decision-making.

Why a Corpus Matters

This is where the concept of a "procurement corpus" becomes important.

A procurement corpus is not simply a collection of documents. It is a structured representation of how procurement decided and acted over time.

It contains not only policies and procedures but also the reasoning behind them. It captures category management approaches, supplier management philosophies, governance models, negotiation strategies, and recurring operational challenges.

Over the last few years, thanks to being the CIPS tutor, I have begun systematically extracting and consolidating IT procurement-related content from multiple sources (training materials, procurement frameworks, category management concepts, automation projects, and endless practical cases I love to share with my course students).

Eventually, I created a reusable knowledge asset.

Once consolidated, an interesting pattern emerged. Certain themes recurred regardless of the specific topic under discussion. The same frustrations also appeared repeatedly.

These recurring patterns became as important as the documents themselves.

Why Procurement Persona Matters Even More

The next realization was that knowledge alone was insufficient.

Even if an AI system has access to every procurement document ever created, it still lacks something important - judgment.

The same information can support multiple decisions depending on the decision-maker's priorities. This led me to begin extracting something different: procurement persona.

The objective was not to model an individual personality. The objective was to capture dominant professional ideas and my specific biases.

A corpus answers the question: “What do I know?” A persona answers the question: “How do I think and judge?” The combination is significantly more powerful than either component individually.

The Evolution Towards HITSS

Long before HITSS existed, I was already teaching IT sourcing strategy as a combination of multiple analytical lenses.

Traditional procurement frameworks provide part of the answer. IT procurement introduces additional considerations. SaaS sourcing introduces another set of variables. Cloud economics, vendor lock-in, governance maturity, and business criticality all influence the eventual sourcing decision.

I never found a single framework that adequately addressed all of these dimensions.

The idea behind Adaptive Sourcing, and later HITSS, was therefore to combine multiple perspectives into a single decision-making framework rather than relying on a traditional methodology.

As part of my Business Automation endeavors, I decided to automate the generation of sourcing strategies, so they incorporate all the different knowledge topics I teach.

I called it a hybrid strategy since it combines many concepts and theories (classical procurement and niche IT ones), each adding extra depth to the eventual analysis and conclusions.

The procurement corpus and the persona model became side products of the automation, which greatly enriched the main output. I tried to get rid of familiar AI language full of truisms.

This created a fundamentally different outcome from generic AI assistants. Instead of producing responses based solely on publicly available information, the system could interpret questions through the lens of my accumulated procurement knowledge and established decision-making principles.

The architecture eventually evolved into what became HITSS.

HITSS Conceptual Architecture

Flowchart illustrating the transformation of personal professional experience and beliefs into actionable decision-support outputs through a structured procurement corpus, persona, HITSS, and comprehensive commercial reasoning processes.

The Future Value Is "Digital Me"

Many discussions about AI focus on model capabilities. I believe this is increasingly the wrong conversation.

Access to capable models is becoming widely available. Over time, model capability itself is likely to become less of a differentiator.

A person who deploys a generic AI assistant gains access to a sophisticated language interface. A person who combines AI with a well-developed corpus and persona gains something far more valuable - reusable personal professional intelligence.

That intelligence already exists in your head. The challenge is not creating it. The challenge is extracting it, structuring it, and making it reusable, effectively creating your professional avatar.

For me, that became the real purpose behind HITSS. It was never about building another AI assistant.

The objective was never to create an AI that replaces procurement professionals. The objective was to capture years of accumulated professional experience, commercial judgment, and sourcing knowledge in a form that remains available, searchable, and reusable.

In simple terms, HITSS is probably more like a digital version of myself than the autonomous AI agents dominating current discussions.

Whether that is useful depends entirely on whether you believe experience still matters.