I have a problem with how to do price discrimination on a product.
I want to find the best pricing strategy, capture customer surplus effectively, but also not damage reseller's business.
Basically I want to have a marketing/pricing strategy that meets the following criteria:
how should I go about devising such a strategy? or should I just have a simple flat price for everyone?
It sounds to me like you're introducing too many variables, too early.
From what you say, right now, you need just three prices to achieve the objectives you describe.
First, you need a price for resellers, [PR] Second, you need a general price, [PG] Third, you need a market testing price, [PT] [PR] might be less, the same, or more than [PG], though most commonly it's going to be x% below [PG] so that the reseller can viably sell at [PG].
[PT] could be more or less than [PG], though from the sound of it you're most likely to want a [PG] that's low enough to attract a broad market and you want to probe higher value opportunities. So you should use [PT] for a test group - and at this stage, I'd just go for a random subset taken from inquiries you feel you can positively identify, because you need to read the potential for value harvesting out of real engagement with your market, rather than writing it into your experiment.
I generally advise people slightly adapt the product, not just the price, for price testing. That introduces some noise, but it also makes clashes more manageable - when you're challenged with, "you quoted me $A, but here's an email from an associate where you quoted $B."
Over time, optimizing prices per market will involve tailoring the proposition, not just the price, if it's to be sustainable - but that's discovery you'll be doing downstream of now.