Not the OP, but how I'd do it: for a v1 there could be some manual outreach and take their word for it, v2 hook into the APIs any ESP provides (probably 2/3 of the big players for this should have outbound reporting APIs), and then the v3 would be a custom pixel that newsletter owners would integrate to be "featured" on the platform.

v4 could then offer manually created ad units, and v5 could be self serve.

Scale is a nice problem to have but I think it is premature to be asking about that now when the business is just trying to get off the ground.

The project was previously featured under the name Mailman on HN [0].

[0] https://news.ycombinator.com/item?id=14107063

I'm still interested in knowing how they're going to do it even at the beginning. Are they going to request access to people's newsletter software for verification or are they just going to take people's word for it?

I would think that Upstart will evolve into a marketplace type of business where the onus is on the business (buyer) to do their due diligence before shelling out money to a newsletter (seller) in the hope of increasing their reach.

Of course if the feature of showing metrics (subscribers, open rates etc) proves to be popular enough they might consider ways to rank newsletters using metrics that are publicly verifiable or internal metrics from running Upstart after several months (that cannot be easily be gamed).

Hopefully @pixelfeeder will respond.

I'd be curious to know how one could do "due diligence" on the size and activity of a private email list. Does that mean contacting people who have advertised in that newsletter before and hoping they share with you their results?

Yes, emailing third parties for social proof is one way to go about it.

Then there's contacting the newsletter directly for proof of their actual numbers by making available current and previous month's screen shots of MailChimp campaign reports, for instance.

And as pixelfeeder mentioned, they'll also gather their own internal analytics based on how money changes hands on Upstart.

Right now, it'll be a lot of manual effort, but you can imagine the data we'll be collecting along the way and eventually being able to rank newsletters and data based on a number of stuff. Including the experience of ad-buyers. With a variety of datapoints + an algorithm to make sense of it, we could eventually (hopefully) rank newsletters based on effectiveness. But I'm no were near that point, so first things first.

Thanks for the response. What does "manual effort" mean? Are you going to get access to their email software to verify? Or maybe they'll send you screenshots?