Skip to content
Exclusive partner BME Symposium18–19 November · Berlin · Booth A16Book a meeting
Contact

A strategy for every category is no longer a headcount question.

Until now, the number of category managers determined how many categories got a real strategy. With codified method, the quality of that method determines it.

In short

A category strategy sets out the levers, suppliers and contract forms procurement uses to steer cost, risk and supply in one category. It rests on an assessment of value, risk, market and feasibility. Until now, usually only the largest categories received a full strategy. With AI and codified method, the same depth becomes possible across the whole portfolio, while the decision stays with the category manager.

Data: third-party research with source. Our view: the joint position of H&Z and Elevatry, indicative observations from project work. Model: calculations and examples, not customer results.

Coverage without depth is dangerous.

A full category strategy easily takes two to three weeks of expert time today. With limited capacity, it gets written for the largest categories. The rest get a template, a carry-forward or nothing. More headcount buys more depth, but only as much as there are positions.

AI can move a procurement team in two ways. Without method, it moves it towards breadth: more categories, more documents, with the reasoning of an average analyst. That looks like progress in the reporting and is risky in a negotiation. With codified method, it moves it towards breadth and depth at once.

The difference shows in capacity. When Companions take on the knowledge-intensive analytical work, and a strategy is updated rather than rebuilt each time, the same expert days cover many times more categories. The model runs on assumptions, not customer data, but the direction is clear: the limit on coverage moves from the number of people to the quality of the method.

Figure 1ModelThe same 180 expert days: all 30 categories actively managed instead of 12, with 60 days left for negotiations.

Categories in the portfolio

12
22
30 of 30
Today: 15 expert days per category, all by handCompanions analyse spend, market and suppliers and draft the strategy: 8 daysCompanions keep every strategy current, your team reviews and decides: 4 days

Model, not customer data. Assumptions: 180 expert days a year for strategy work, 30 categories in the portfolio. Companions take on data preparation, assessment, market research and the first strategy draft (8 instead of 15 days). With stored assumptions and continuous market monitoring, about 4 days per category remain for review and decision: 30 × 4 = 120 days, 60 days free.

Good strategy starts with criteria, not with a template.

What separates an experienced category manager from a template is the way she reads a category. She looks at value: profit impact, negotiating leverage, index exposure. At risk: supply, supplier concentration, compliance. At market behaviour, demand volatility and switching costs. And at what can realistically change: standardisation, substitution, make-or-buy.

These criteria are rarely written down. They live in experience. Elevatry and H&Z, drawing on more than 5,000 procurement projects, have carried them into the Companions. They assess every category against the same criteria, derive a strategy and show which assessment led to which recommendation. The four lenses bundle the twelve weighted criteria of the H&Z methodology; the position in the Kraljic portfolio follows from value and supply risk.

That changes three things. The hundredth category is analysed to the same standard as the largest. A new buyer follows the logic of the best colleagues in her first month, and sees why. And every recommendation can be checked, because criteria and evidence are visible.

Figure 2ModelHigh value, manageable risk, good feasibility: the assessment points to bundling and standardising.
Value: what is at stakehigh = large lever4.3 of 5
Risk: what could hurthigh = critical2.3 of 5
Market: supplier powerhigh = supplier holds the cards2.0 of 5
Feasibility: what can changehigh = easy to implement3.7 of 5

Example assessment of an electric motors category, scale 1 (low) to 5 (high), criteria based on H&Z category strategy methodology. Scores illustrative.

Strategies start ageing the day they are approved.

A category strategy rests on assumptions: raw material prices, market capacity, the supplier landscape, demand. Once approved, those assumptions begin to shift. An index turns, a supplier is acquired, a plant moves volume. In the folder, the strategy stays the same until someone has time to revise it.

Then there is the loss of knowledge. The reasoning behind a strategy often lives in the heads of the people who built it. When someone leaves, the document stays and the logic goes. The next revision starts from zero. The same questions get asked again, the same data searched for again.

When a strategy is stored in the Company Brain with its assumptions and reasons, Companions can watch those assumptions. If one changes measurably, the system flags it with the consequence in euros and a proposed adjustment. The strategy becomes a living working state, and approving the change stays with the category manager.

50–80%of expert time goes into knowledge-intensive, complex preparation work.Our viewIndicative observation from H&Z and Elevatry project work, not a representative benchmark
A strategy whose assumptions nobody watches is a document, not a strategy.

Companions for this topic

What Elevatry does for you.

Companions work around the clock in your data, with the H&Z method from more than 5,000 procurement projects. They prepare decisions, work in your experts' feedback and, once you approve, carry out the next steps. Your team decides.

All Companions
Model

A strategy for electric motors nobody had touched in four years

A plant engineering company buys electric motors worth €6.5M a year for three plants from eleven suppliers. The last strategy is four years old. Two suppliers account for 68% of volume.

  1. 01
    Read spend and contracts

    The Companion finds 14 motor variants for applications that six could cover. One framework agreement applies the copper surcharge to the full motor price rather than to the copper share.

  2. 02
    Assess through four lenses

    High value and high index exposure, moderate supply risk, good potential for standardisation. The market offers several qualified suppliers within delivery range.

  3. 03
    Derive the strategy

    Reduce variants from 14 to six, bundle volume across plants, limit the copper formula to the material share and keep a second source for resilience.

  4. 04
    Put a number on it

    Standardisation and bundling are worth about 3.5% of volume, roughly €0.23M. The corrected copper formula adds about €0.07M a year.

  5. 05
    The category manager decides

    Whether engineering supports the variant reduction, and how fast to switch, is settled by a person with engineering and the plants.

About €0.3M a year from a category without a current strategy, prepared in around eight expert days instead of fifteen.

Illustrative, not a customer result.

Questions worth asking.

  1. 01

    For how many of our categories is there a strategy less than two years old?

  2. 02

    Would the strategy for our hundredth category pass the same review as the one for our largest?

  3. 03

    Where is our method written down so that a new category manager can follow it?

  4. 04

    Which assumption in our approved strategies is now wrong, and who would notice?

What procurement leaders ask us.

How much can AI realistically save in category strategy?
In our worked example, about 4.6% of volume, but the larger lever is how many categories have a strategy at all. For electric motors with €6.5M of spend, cutting variants from 14 to six, bundling across three plants and limiting the copper formula to the material share yield about €0.3M a year. The last strategy was four years old. Your potential is therefore not a percentage but the number of categories without a maintained strategy.
How long does a category strategy take with AI?
In our worked example, about eight expert days instead of fifteen, and around four in the following year in our model. Companions take on the knowledge-intensive analytical work that makes up most of a strategy today. And a strategy with stored assumptions is updated rather than rebuilt each time. What matters is not the speed of one strategy but what it frees up: the same expert days cover far more categories, and the reasoning survives the next change of staff.
How many categories can a procurement team cover strategically with AI?
In our model, all 30 categories actively managed instead of 12 with the same 180 expert days a year, with 60 days to spare. Assumptions: Companions take on analysis, market research and the strategy draft, and with continuous monitoring a category costs about four expert days a year for review and decision. Until now, the number of category managers decided how many categories got a real strategy. From now on, the quality of the method decides. Codified method takes it across the whole portfolio.
Does an AI-generated category strategy hold up to expert review?
Yes, if it rests on criteria rather than a template. AI without method writes strategies with the reasoning of an average analyst. That reads as progress and fails in negotiation. Elevatry assesses every category against criteria from over 5,000 H&Z projects, covering value, risk, market behaviour and feasibility, and shows which assessment led to which recommendation. The hundredth category is then held to the same standard as an A category. Which levers to pull is the category manager's decision.
How do I keep category strategies current without rewriting them every year?
Monitor the assumptions, not the document. Every strategy rests on raw material prices, market capacity, the supplier landscape and demand, and these start shifting the day it is approved. An index turns, a supplier is acquired, a plant moves volume. When assumptions and reasoning are stored with the strategy, the system flags a measurable shift with its consequence in euros and a proposed adjustment. The annual document becomes a living working state. Approval stays with the category manager.

Friday notes on the topic

Next step

See it on your own procurement.

We show you on one of your processes how Companions do the work from analysis to contract. With your data and your approval rules, if you like.

See the platform