A member company with five million euros in revenue sources around 820,000 euros of indirect demand per year, narrowly defined as energy plus purchased services and consulting. At two million euros in revenue it is around 330,000 euros. Twenty member companies with five million each come to around 16 million together, fifty to around 41 million.
Those totals look substantial. For the question of whether enough volume comes together, they are the wrong measure.
What gets tendered is a category, not a total volume.
Assuming fifteen bundleable categories with roughly equal distribution, a different picture emerges.
A single member company reaches around 55,000 euros per category per year. Twenty member companies together reach around 1.1 million. Fifty reach around 2.7 million.
The split across fifteen categories is an assumption for illustration, not a survey. With ten categories the amounts per category rise by half. That does not change the order of magnitude.
This is the actual point. An order worth 55,000 euros a year does not justify the effort of a structured procedure for anyone involved.
The buyer at the member company renews the existing contract, because three days of work cannot be justified for that amount. The supplier who would have to bid prices it carelessly or does not bid at all. A careful bid costs working time, and below a certain order size that effort does not pay.
Above one million euros the same supplier behaves differently. Not out of goodwill, but because the effort now pays and because they have to assume the other invitees are calculating the same way.
So in this category bundling does not only produce a better price. It creates the market in the first place.
In the core assortment the question does not arise with the same sharpness. A retail company with five million euros in revenue already has volumes there that interest a supplier, and in a buying group these assortments are usually covered by centrally negotiated terms.
With indirect demand it is the other way around. It spreads across many small categories, none of which reaches a size at the individual member that would carry a tender, and in total it makes up a considerable share of addressable costs.
Many small individual volumes with a relevant total. That is precisely the constellation bundling is made for.
Not every requirement suits a joint tender. Three groups need to be distinguished.
Well suited to bundling is standardizable indirect demand that arises at all members in comparable form. This includes energy, packaging material, workwear and personal protective equipment, cleaning and waste disposal services, telecommunications, office supplies and part of maintenance and logistics services.
Conditionally bundleable are services with regional ties or individual specification. Trade services, maintenance of site-bound installations or insurance with widely differing risk profiles can be tendered jointly, but lead to lots rather than a single award.
Not bundleable are site-bound cost positions such as rent and leases, and as a rule the core assortment where centrally negotiated terms already exist. Central settlement, meaning the bundled handling of payment flows, is also not bundling in the sense meant here. It concerns invoicing, not awarding.
How large is the field overall? The German federal association of cooperating SMEs represents, by its own account, around 250 buying groups with roughly 200,000 affiliated companies and combined revenue of more than 513 billion euros. Arithmetically that is an average of around 800 members per buying group. That average is heavily shaped by a few very large cooperations and describes the typical case poorly. This is why the calculations above start at the individual member company.
Our members differ greatly in size. Does the calculation still work? Yes, and size differences tend to help. What matters is the total per category, not equal distribution. One large member can lift a category across the threshold at which a supplier prices carefully.
Do we have to start with all fifteen categories? No. Participation is decided per category, and experience shows the start is easier with two or three categories where the specification is uncontroversial.
Chapter 6 of the study contains the full bundling calculation with every intermediate step, the taxonomy of the three bundling groups and the section on who carries the effort in a buying group and who receives the benefit.
→ To the white paper
Those totals look substantial. For the question of whether enough volume comes together, they are the wrong measure.
What gets tendered is a category, not a total volume.
The calculation that matters
Assuming fifteen bundleable categories with roughly equal distribution, a different picture emerges.
A single member company reaches around 55,000 euros per category per year. Twenty member companies together reach around 1.1 million. Fifty reach around 2.7 million.
The split across fifteen categories is an assumption for illustration, not a survey. With ten categories the amounts per category rise by half. That does not change the order of magnitude.
Why 55,000 euros is too little
This is the actual point. An order worth 55,000 euros a year does not justify the effort of a structured procedure for anyone involved.
The buyer at the member company renews the existing contract, because three days of work cannot be justified for that amount. The supplier who would have to bid prices it carelessly or does not bid at all. A careful bid costs working time, and below a certain order size that effort does not pay.
Above one million euros the same supplier behaves differently. Not out of goodwill, but because the effort now pays and because they have to assume the other invitees are calculating the same way.
So in this category bundling does not only produce a better price. It creates the market in the first place.
Why the effect appears with indirect demand
In the core assortment the question does not arise with the same sharpness. A retail company with five million euros in revenue already has volumes there that interest a supplier, and in a buying group these assortments are usually covered by centrally negotiated terms.
With indirect demand it is the other way around. It spreads across many small categories, none of which reaches a size at the individual member that would carry a tender, and in total it makes up a considerable share of addressable costs.
Many small individual volumes with a relevant total. That is precisely the constellation bundling is made for.
What can be bundled and what cannot
Not every requirement suits a joint tender. Three groups need to be distinguished.
Well suited to bundling is standardizable indirect demand that arises at all members in comparable form. This includes energy, packaging material, workwear and personal protective equipment, cleaning and waste disposal services, telecommunications, office supplies and part of maintenance and logistics services.
Conditionally bundleable are services with regional ties or individual specification. Trade services, maintenance of site-bound installations or insurance with widely differing risk profiles can be tendered jointly, but lead to lots rather than a single award.
Not bundleable are site-bound cost positions such as rent and leases, and as a rule the core assortment where centrally negotiated terms already exist. Central settlement, meaning the bundled handling of payment flows, is also not bundling in the sense meant here. It concerns invoicing, not awarding.
Frequently asked questions
How large is the field overall? The German federal association of cooperating SMEs represents, by its own account, around 250 buying groups with roughly 200,000 affiliated companies and combined revenue of more than 513 billion euros. Arithmetically that is an average of around 800 members per buying group. That average is heavily shaped by a few very large cooperations and describes the typical case poorly. This is why the calculations above start at the individual member company.
Our members differ greatly in size. Does the calculation still work? Yes, and size differences tend to help. What matters is the total per category, not equal distribution. One large member can lift a category across the threshold at which a supplier prices carefully.
Do we have to start with all fifteen categories? No. Participation is decided per category, and experience shows the start is easier with two or three categories where the specification is uncontroversial.
The next step
Chapter 6 of the study contains the full bundling calculation with every intermediate step, the taxonomy of the three bundling groups and the section on who carries the effort in a buying group and who receives the benefit.
→ To the white paper