I think your on-demand delivery to a GPS coordinate idea is the best part of this!
I'm not saying don't do the marketplace thing, but if "40% of Online stores in Myanmar loses sales because the delivery person can't find address of the customer" is true, then you have found an amazing market to tap.
Are you based in Myanmar? Can you describe what it's like trying to get something delivered?
Is this a mapping problem, or a city organization/government problem? How many streets are missing from maps? If OpenStreetMap or Google decided to improve the maps would the on-demand delivery market still exist?
Here's what I'm picturing: A retailer has their normal store, either with WhatsApp, Facebook, or maybe even with you. They include a button that says "Buy with Esobo" which kicks the user into your app. The user gives you their payment information and when they click "confirm" your app records their current location (or desired location and time in the future) and notifies the drivers who have signed up to deliver packages. A driver clicks "confirm", drives to the store, picks up the package, and then drives to the user's location, shown as a point on a Map. You'll be relying on drivers' knowledge of the city for better navigation that you can get from the latest maps. You charge a little bit to pay the drivers, you keep a bit for yourself, the retailers are happy because they made a sale, and the users are happy because they got what they wanted for a small delivery fee.
I think this could be absolutely huge in countries without a strong centralized delivery infrastructure.