Salespeople don’t lie, but they do lean the boat into the wind which suits them best, and shear away from facts that are in the wrong quarter, writes John Gadd. So how can the buyer distinguish selling fact from optimistic fiction?
Last time, I outlined the value on the Return on Extra Outlay (REO) concept to help assess the value of a product’s claims. Farmers often – and rightly – are a rather cynical lot, and point out that it is the “returns” aspect of the equation that influences it the most, and they are right to do so. And while not a cast-iron solution, a few pointed questions can help sort the men from the boys in this respect.
Seven key questions
I am obliged to world-renowned Dr Gordon Rosen and two other statisticians for the following list of suggested questions to ask your supplier. The first four to screen a seller’s or manufacturer’s claims will reveal those that are questionable and rejectable. If undecided, the remaining three should be referred to a statistician or trials officer at a local college or animal research station – for a fee of course.
“Too expensive” goes up the cry, and this I will address next time, as I’ve used them from time to time without my modest resources being wrecked!
The salesperson may well not be able to answer these questions themselves, but they can refer them to their company. Sometimes the product might be too new to have had the time do provide sufficient trial results, and this can be taken into consideration. But it’s the availability, or not, of answers to the first four that will sort out the constructive men from the over-optimistic boys.
There are usually plenty of alternative products out there, so move on if the answers seem unsettling. These questions are best suited to feeds or feed supplements.
- How many properly controlled tests do you have for the product?
Dependent on numbers of pigs involved, about 20 are needed.
- How many of these tests had no negative controls?
All trials need a negative control.
- Have you a list of references to support the above?
Needed to show whether the trials were in-house or done by an independent source. 50% independent sources is advisable.
- How many times out of 10 did the product improve performance?
Seven is acceptable; 10 is highly suspicious!
- What’s the response coefficient of variation?
These reveal the amount of variation in responses over a range of conditions, as your farm isn’t the same as anybody else’s. I’m told variations of up to 50% are acceptable.
- Can you supply a dose response curve?
Preferable to have this so as to get best value for money. So ask what is the predicted return on investment at half and again double the recommended level.
- Have you models to predict responses from your product under my conditions?
Enables comparisons to be made with alternative investments.