Is trade promo failure a CPG thing?

We frequently hear numbers on the failure of trade promotions indicating that 70% or more of promotional events are unsuccessful for the manufacturer (though not necessarily the retailer), even with "success" being rather limply defined as producing enough lift to pay for the cost of the promotion. I've also heard it said that the average promotion returns 65c for every dollar spent.

Ouch. If I really believed that I worked in a business whose function was to subtract value, I'd look for another trade. The following is an example of results I've often seen, based on supplier and retailer promo profitability:

What I find interesting is that these horrific numbers always come from the CPG side of the trade promo world. Over in durables and business-to-business, the story is very different.

This particular graph comes from the hardware/d-i-y world and represents sales growth results for retailers based on the level of their usage of a particular manufacturer's co-op/mdf program. In an earlier life, I did this same study for probably a hundred manufacturers, in a variety of durables and B2B categories, and the results were similarly positive in all but two or three cases.

I can think of four possible explanations for these apparently contradictory results:

1) Trade promotion doesn't work: The CPG numbers are true, the durables/B2B numbers false -- a reflection of the more sophisticated analytics available in CPG.

2) Trade promotion works: The durables/B2B numbers are true, the CPG numbers somewhat false, because in CPG many promotions that are planned and paid for don't actually happen. Therefore the negative numbers in CPG are often the result not of unsuccessful promotions, but of promotions that never happened.

3) Trade promotion works if it's done right: Both sets of numbers are true, and are reflective of the fact that durables and B2B trade promo programs have more in the way of rules and structure behind them, which leads to a more disciplined usage of trade promo funds. This is somewhat related to #2, since in durables/B2B, part of the rules/structure I mentioned is that documentation of events is generally required.

4) Maybe it works and maybe it doesn't: Both sets of numbers are false -- which means we don't have a clue what's going on.

I have no way of proving any of these contentions, but my belief is that #3 is closest -- that trade promotion, when used in a targeted manner to promote to the end-buyer, and with the discipline of documentation behind it, generally drives increased sell-through.

The question for CPG manufacturers is how to re-introduce discipline in their programs. For durables/B2B manufacturers, who seem to be moving quickly toward the CPG model, the question is how to avoid following CPG over the cliff. I don't pretend to know for sure, but I suspect the answers lie in better analytics and better use of the resulting findings.