I'm not going to judge the maker one way or the other since I don't have data to back it up either way. I can only speak for some of the different carbs we've worked with, mostly side draft, motorcycle, aircraft, etc that experienced poor burn efficiency when introducing a stream of fuel with every roll of the throttle, as opposed to breaking that stream up in some manner. Manifold designs, flow, and post fuel delivery turbulence were far less effective than changing the patterns in which a fuel mixture was introduced. The improvements were significant when viewed via a wide band sensor during dyno runs. I agree you won't likely see MPG improvements when concentrating on this pump, unless under constant stop and go traffic conditions requiring a constant injection from the accelerator pump.
It would probably be better to get that company to weigh in with data rather than using anecdotal evidence or comparisons (mine included
)
As Deming's old saying goes, "In God we trust, all others must bring data." Neither a claim of success or statements that something will not work is of no value unless data is used to back it up either side.
My recommendations to the OP if interested in this:
1) Speak with the company about any guarantee
2) Run back to back test with and without the device under identical driving conditions.
3) Report the findings so that the device's value can truly be judged.