![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
On not outsmarting the customer
Mar. 14th, 2013 09:29 pmI am still on the road with semi-broken laptop, periodically reading LJ but I don't have time for posting yet.
Last week I was in a meeting with a customer together with few colleagues. One of my colleagues is a senior principal engineer. She is a very bright person, and she knows everything about CPUs inside out. When I was observing her communication with customer's engineers, I found a very interesting difference between how she is doing it and how I talk with customer. I think her approach makes more sense than mine.
Let's assign a technical depth level to a CPU architecture question customer wants to discuss. E.g. starting from 1 - "MMX speeds up internet, the more Ghz the better, multicore is cool, etc." up to 10, discussing actual RTL models. Then there are three variables on this scale related to the discussion: C == level of current customer's understanding, P == level that is actually required for meaningful discussion, and last but not least - E == level of support engineer's understanding. (There is also level of NDA that is the ultimate cap, but it is irrelevant for the case I am describing).
My approach was always the following: if my E < P, I look for an expert and refer a customer to this expert. If my E >= P, I start discussion at E. If I see it is too complicated for a customer, I reduce E one by one, until I reach max(C, P). If P > C, I say so as polite as I can and point to the sources of information that may help to get C up to P.
H.'s approach is different. She does not demonstrate her expert knowledge, everyone knows that her E=[9-10] (comparing to mine which is [4-7]). Nevertheless, she always starts discussion at a level which equals min(C-1, P). Then she may go up if customer requests so, but still never exceeds min(C, P). This makes more sense, does it? I guess with bringing higher E to discussion I am just showing off?
Last week I was in a meeting with a customer together with few colleagues. One of my colleagues is a senior principal engineer. She is a very bright person, and she knows everything about CPUs inside out. When I was observing her communication with customer's engineers, I found a very interesting difference between how she is doing it and how I talk with customer. I think her approach makes more sense than mine.
Let's assign a technical depth level to a CPU architecture question customer wants to discuss. E.g. starting from 1 - "MMX speeds up internet, the more Ghz the better, multicore is cool, etc." up to 10, discussing actual RTL models. Then there are three variables on this scale related to the discussion: C == level of current customer's understanding, P == level that is actually required for meaningful discussion, and last but not least - E == level of support engineer's understanding. (There is also level of NDA that is the ultimate cap, but it is irrelevant for the case I am describing).
My approach was always the following: if my E < P, I look for an expert and refer a customer to this expert. If my E >= P, I start discussion at E. If I see it is too complicated for a customer, I reduce E one by one, until I reach max(C, P). If P > C, I say so as polite as I can and point to the sources of information that may help to get C up to P.
H.'s approach is different. She does not demonstrate her expert knowledge, everyone knows that her E=[9-10] (comparing to mine which is [4-7]). Nevertheless, she always starts discussion at a level which equals min(C-1, P). Then she may go up if customer requests so, but still never exceeds min(C, P). This makes more sense, does it? I guess with bringing higher E to discussion I am just showing off?
