A paper that offers a solution for every case helps in none. That is why our study names five constellations in which a tendering tool does not pay off on the numbers.
We sell the kind of tool whose effect is examined there. The five points stand without qualification anyway.
The bundleable volume is too small. As an orientation it should be twenty-five to sixty-seven times annual total cost, depending on the assumptions. If it sits well below that, the investment does not pay even under favourable assumptions.
The current tender rate is already high. Whoever awards eighty percent of their volume in structured procedures has little new volume to lift. The remaining effect is limited to improvement within the existing base and is correspondingly small.
Demand is complex and changes frequently. Bajari, Tadelis and McMillan find, in an empirical study of procurement procedures, that negotiations outperform competitive tendering where the service is complex and substantial subsequent changes are to be expected. The reason is that a competitive procedure depends on a fixed specification.
The supplier field is narrow. Where only two or three suppliers come into question, the mechanisms of a structured procedure work only in a limited way. A supplier's mark-up responds to the expected number of competitors, and that number is small here.
Participation cannot be organized. A bundle that works on paper but that nobody joins is not a bundle.
The second point deserves a note.
An organization with a high tender rate and experienced buyers is the hardest case for a business case, not the easiest. The greatest benefit arises where little has been awarded in structured form so far, and that is rarely the organization that asks for a tool first.
For buying groups this means the following. A central office that already brings its fifteen most important categories to market every year will gain less through volume than one where half the contracts are quietly renewed for lack of time. The second has the worse starting position and the better business case.
Experience shows that projects in buying groups do not fail on economics. They fail on consent.
The numbers can work, the volume can suffice, the categories can fit, and still no procedure takes place, because the member companies do not report their demand or participation is not decided internally.
That is not an arithmetic given and therefore belongs in the decision. Anyone who cannot organize consent should postpone the purchase rather than dress up the calculation.
The path there is shorter than it looks. The jump that matters lies between one participant and a double-digit number of participants, and participation can be decided per category. Starting with two or three uncontroversial categories and twenty member companies is something other than a decision of principle for everyone.
It does not follow that a tool has no value in these cases. The five constellations look exclusively at cash-effective savings, meaning the part that can be expressed in euros.
An organization that already awards in structured form may use a tool for reasons that appear in no business case. Award decisions become evidenceable towards auditors, governing bodies, auditing associations and funding providers. The separation of information between legally independent participants can be established technically and evidenced afterwards. Procedures become independent of individuals and survive staff turnover. And the central office of a buying group gains a structured overview of its members' actual demand.
In our calculations these points are set at zero. That does not make them smaller, it only makes them unquantifiable.
How do we quickly check which case we are in? With two numbers. The bundleable volume of the participating member companies in the categories actually intended for joint awarding, and the share of that already awarded in structured form today. The rest is fine calculation.
We have one complex category and fifteen simple ones. What now? The question arises per category, not for the organization. The third and fourth points exclude individual categories, not the project.
Why is this not on your product page? Fair enough. It is in the study, and it is here now.
Chapter 8 of the study sets out the five constellations with their rationale, the break-even calculation under three sets of assumptions and the orientation figure for a first check.
→ To the white paper
We sell the kind of tool whose effect is examined there. The five points stand without qualification anyway.
The five cases
The bundleable volume is too small. As an orientation it should be twenty-five to sixty-seven times annual total cost, depending on the assumptions. If it sits well below that, the investment does not pay even under favourable assumptions.
The current tender rate is already high. Whoever awards eighty percent of their volume in structured procedures has little new volume to lift. The remaining effect is limited to improvement within the existing base and is correspondingly small.
Demand is complex and changes frequently. Bajari, Tadelis and McMillan find, in an empirical study of procurement procedures, that negotiations outperform competitive tendering where the service is complex and substantial subsequent changes are to be expected. The reason is that a competitive procedure depends on a fixed specification.
The supplier field is narrow. Where only two or three suppliers come into question, the mechanisms of a structured procedure work only in a limited way. A supplier's mark-up responds to the expected number of competitors, and that number is small here.
Participation cannot be organized. A bundle that works on paper but that nobody joins is not a bundle.
The point that runs counter to our sales interest
The second point deserves a note.
An organization with a high tender rate and experienced buyers is the hardest case for a business case, not the easiest. The greatest benefit arises where little has been awarded in structured form so far, and that is rarely the organization that asks for a tool first.
For buying groups this means the following. A central office that already brings its fifteen most important categories to market every year will gain less through volume than one where half the contracts are quietly renewed for lack of time. The second has the worse starting position and the better business case.
In practice the fifth point is the most common
Experience shows that projects in buying groups do not fail on economics. They fail on consent.
The numbers can work, the volume can suffice, the categories can fit, and still no procedure takes place, because the member companies do not report their demand or participation is not decided internally.
That is not an arithmetic given and therefore belongs in the decision. Anyone who cannot organize consent should postpone the purchase rather than dress up the calculation.
The path there is shorter than it looks. The jump that matters lies between one participant and a double-digit number of participants, and participation can be decided per category. Starting with two or three uncontroversial categories and twenty member companies is something other than a decision of principle for everyone.
What a tool delivers even in these cases
It does not follow that a tool has no value in these cases. The five constellations look exclusively at cash-effective savings, meaning the part that can be expressed in euros.
An organization that already awards in structured form may use a tool for reasons that appear in no business case. Award decisions become evidenceable towards auditors, governing bodies, auditing associations and funding providers. The separation of information between legally independent participants can be established technically and evidenced afterwards. Procedures become independent of individuals and survive staff turnover. And the central office of a buying group gains a structured overview of its members' actual demand.
In our calculations these points are set at zero. That does not make them smaller, it only makes them unquantifiable.
Frequently asked questions
How do we quickly check which case we are in? With two numbers. The bundleable volume of the participating member companies in the categories actually intended for joint awarding, and the share of that already awarded in structured form today. The rest is fine calculation.
We have one complex category and fifteen simple ones. What now? The question arises per category, not for the organization. The third and fourth points exclude individual categories, not the project.
Why is this not on your product page? Fair enough. It is in the study, and it is here now.
The next step
Chapter 8 of the study sets out the five constellations with their rationale, the break-even calculation under three sets of assumptions and the orientation figure for a first check.
→ To the white paper