The debate in the procurement industry revolves around whether eight or fifteen percent savings are achievable. For a managing director looking at the annual result, that question is secondary. The reason is a reversal of the usual calculation.
The usual question is what a five percent price reduction delivers. The better question is how small the price reduction can be for the annual result to improve by five percent.
For a retail company it is 1.2 percent.
The calculation needs two numbers, and both come from the same source, the KfW SME Panel 2025. The first is the pre-tax return on sales, the second the share of indirectly sourceable spend in total costs.
For retail the return on sales is 3.8 percent. Of the cost structure, 11 percent goes to energy and 6 percent to purchased services and consulting, 17 percent together. Both positions are tenderable and can be bundled across several units.
Revenue cancels out in the calculation. It is therefore independent of company size and requires neither a figure for procurement volume nor its share of revenue.
A five percent price reduction on this part of costs produces a profit improvement of 21.5 percent in retail, 13.7 percent in manufacturing, 9.3 percent in construction and 12.6 percent across the mid-market as a whole.
In a buying group, negotiation is central and procurement is local. The central office runs the process, the member company receives the goods and pays the invoice.
The profit effect therefore arises there, at the individual member, and not at the central office. For the argument towards the member companies that is the decisive figure. Not what the group moves in total, but what lands in their own accounts.
And there the calculation meets an uncomfortable property of retail. The margin is thin. That is exactly what magnifies the lever. The lower the return on sales, the greater the relative effect of the same absolute saving. Sectors with thin margins benefit most from an improvement in purchasing, and retail is the sector with the thinnest margin.
More informative than asking what five percent achieves is the reverse question.
Related to the return on sales of the individual sectors, the picture looks like this. A retail company needs around 1.2 percent price reduction for five percent more profit. Manufacturing needs 1.8 percent, the mid-market as a whole 2.0 percent, construction 2.7 percent.
1.2 percent lies within the spread of a single bid. That is not a negotiation success anyone writes a report about. Whether the industry promises eight or fifteen percent is secondary to this effect.
Even in the strictest variant the statement holds. If energy is left out entirely, because many companies already award it in structured form and on fixed cycles, a retail company needs a 3.3 percent price reduction on purchased services for the same five percent profit improvement. That is still well below any industry figure.
The obvious approach would be to base the calculation on cost of goods, because that is where the largest volume sits. Two reasons argue against it.
The first is pass-through to the end customer. A retail company that lowers its purchase prices passes the advantage on in competition, wholly or partly, through its selling price. How large that share is depends on competitive intensity in the respective category. For indirect material the question does not arise, because it does not enter the selling price of the goods.
The second is a plausibility check. In retail, 38 percent of total costs fall on material, raw materials and intermediate products. A five percent price reduction on that would produce, at a 3.8 percent return on sales, a profit improvement of around 48 percent. Doubling the annual result through a five percent price reduction does not describe anything observed in practice.
One of two things follows. Either the pass-through rate is high, or the widespread industry figures never referred to the entire cost of goods. Both conclusions lead to the same consequence, namely to base the calculation where it holds.
For buying groups this matches practice. What gets bundled is usually indirect and peripheral demand, while the core assortment is already covered by existing terms.
Does the calculation apply to our company? The method does, the sector values not necessarily. The return on sales in the panel is a revenue-weighted average in which larger companies carry more weight. Anyone applying the calculation to their own company should use their own return on sales and their own cost share.
What if five percent is too optimistic for us? The profit effect is strictly proportional. Half the value halves the result. At 2.5 percent the effect in retail is still around 10.8 percent.
Do you include process costs? No. Saved working time only changes the annual result if it actually leads to lower personnel costs, and in the mid-market that is the exception. The freed capacity goes into additional tenders, and that effect is already captured elsewhere.
Chapter 5 of the study contains the full data basis, the calculation path as a formula, three differently wide definitions of indirectly sourceable spend and the limits of the calculation in one place.
→ To the white paper
The usual question is what a five percent price reduction delivers. The better question is how small the price reduction can be for the annual result to improve by five percent.
For a retail company it is 1.2 percent.
The calculation with two inputs
The calculation needs two numbers, and both come from the same source, the KfW SME Panel 2025. The first is the pre-tax return on sales, the second the share of indirectly sourceable spend in total costs.
For retail the return on sales is 3.8 percent. Of the cost structure, 11 percent goes to energy and 6 percent to purchased services and consulting, 17 percent together. Both positions are tenderable and can be bundled across several units.
Revenue cancels out in the calculation. It is therefore independent of company size and requires neither a figure for procurement volume nor its share of revenue.
A five percent price reduction on this part of costs produces a profit improvement of 21.5 percent in retail, 13.7 percent in manufacturing, 9.3 percent in construction and 12.6 percent across the mid-market as a whole.
Why the effect arises at the member company
In a buying group, negotiation is central and procurement is local. The central office runs the process, the member company receives the goods and pays the invoice.
The profit effect therefore arises there, at the individual member, and not at the central office. For the argument towards the member companies that is the decisive figure. Not what the group moves in total, but what lands in their own accounts.
And there the calculation meets an uncomfortable property of retail. The margin is thin. That is exactly what magnifies the lever. The lower the return on sales, the greater the relative effect of the same absolute saving. Sectors with thin margins benefit most from an improvement in purchasing, and retail is the sector with the thinnest margin.
The reversal that matters
More informative than asking what five percent achieves is the reverse question.
Related to the return on sales of the individual sectors, the picture looks like this. A retail company needs around 1.2 percent price reduction for five percent more profit. Manufacturing needs 1.8 percent, the mid-market as a whole 2.0 percent, construction 2.7 percent.
1.2 percent lies within the spread of a single bid. That is not a negotiation success anyone writes a report about. Whether the industry promises eight or fifteen percent is secondary to this effect.
Even in the strictest variant the statement holds. If energy is left out entirely, because many companies already award it in structured form and on fixed cycles, a retail company needs a 3.3 percent price reduction on purchased services for the same five percent profit improvement. That is still well below any industry figure.
Why we do not calculate on cost of goods
The obvious approach would be to base the calculation on cost of goods, because that is where the largest volume sits. Two reasons argue against it.
The first is pass-through to the end customer. A retail company that lowers its purchase prices passes the advantage on in competition, wholly or partly, through its selling price. How large that share is depends on competitive intensity in the respective category. For indirect material the question does not arise, because it does not enter the selling price of the goods.
The second is a plausibility check. In retail, 38 percent of total costs fall on material, raw materials and intermediate products. A five percent price reduction on that would produce, at a 3.8 percent return on sales, a profit improvement of around 48 percent. Doubling the annual result through a five percent price reduction does not describe anything observed in practice.
One of two things follows. Either the pass-through rate is high, or the widespread industry figures never referred to the entire cost of goods. Both conclusions lead to the same consequence, namely to base the calculation where it holds.
For buying groups this matches practice. What gets bundled is usually indirect and peripheral demand, while the core assortment is already covered by existing terms.
Frequently asked questions
Does the calculation apply to our company? The method does, the sector values not necessarily. The return on sales in the panel is a revenue-weighted average in which larger companies carry more weight. Anyone applying the calculation to their own company should use their own return on sales and their own cost share.
What if five percent is too optimistic for us? The profit effect is strictly proportional. Half the value halves the result. At 2.5 percent the effect in retail is still around 10.8 percent.
Do you include process costs? No. Saved working time only changes the annual result if it actually leads to lower personnel costs, and in the mid-market that is the exception. The freed capacity goes into additional tenders, and that effect is already captured elsewhere.
The next step
Chapter 5 of the study contains the full data basis, the calculation path as a formula, three differently wide definitions of indirectly sourceable spend and the limits of the calculation in one place.
→ To the white paper