Native Outreach, and we're our own first customer
We built a full outbound engine inside Native. Lead research, AI drafting that won't invent facts, and sending from your own inbox. We're our own first customer.
Ten days since the last note. This one is mostly about a single big thing we built, a smaller thing that will quietly make your evenings better, and a look at where Native is heading next.
Native Outreach
Outbound email is the one job people still leave Native to do. You find people somewhere, draft in another tool, paste into a sending tool, and stitch the replies back together by hand. So we built the whole thing in.
- Tell it who you want to reach. Describe your ideal customer, paste a list, upload a CSV, or point it at your own contacts. Native goes and finds the people (from the web, from Apollo, from whatever you gave it), checks that each one has a real, usable email, and dedupes them against everyone you already know so you are not emailing the same person twice.
- It drafts like a person, and it will not make things up. Native personalizes the subject line and the opening and closing of every email to the recipient. The hard rules matter more than the personalization: it is forbidden from inventing a fact about someone, from claiming a mutual connection that does not exist, and from reaching for "synergy" or "revolutionary" or any of the filler that makes cold email read like cold email. Under a hundred and fifty words. If it is ever unsure, it falls back to your own template instead of guessing.
- It sends from your own inbox. Gmail or Outlook, one to one, so the person on the other end sees a real human and their reply lands in your thread like any other email. No tracking pixel on these. A warm send stays warm.
- It is built to protect your sending reputation. Native paces sending per inbox and per recipient domain to lower the odds of tripping a spam filter. Bounces are suppressed automatically. The moment someone replies, the follow-ups to that person stop. Nobody gets a robotic "just bumping this" after they have already answered. And anyone who asks to be left alone is suppressed for good, so a campaign never reaches them again.
- Nothing sends without you. Every message waits in an approval queue. You read it, you approve it, then it goes. The AI drafts; you decide.
Now the part I want to be straight about. Outreach is not switched on for everyone yet. Over the next few days we are pointing it at our own outbound. Native is going to be its own first customer. We will run real campaigns on it, watch what breaks, and fix it where it bleeds. And then, because building in public only counts if you show the receipts, I will come back with the actual numbers: how many we sent, what landed, what replied. No vanity slides.
If you want in on the first group, reply to this. I will add you.
Your daily digest got a quiet glow-up
If you are on Gmail, Native already sends you one digest each evening of the low-priority mail it kept out of your way: the newsletters, the receipts, the notifications that do not need to interrupt your day. Two things changed.
- It actually arrives now. Once. The old version had a bad habit. Some evenings you got two, some evenings none, and occasionally it would get wedged and stop coming altogether. That is gone. One digest, every evening, and if the first attempt slips it quietly recovers later the same night instead of giving up.
- It reads better, and it pulls you back in. Every item is now a clean card with the suggested move (Archive, Unsubscribe, Mark as spam) sitting right there as a chip you can read at a glance. And a single button opens your inbox inside Native, so you can clear the whole pile in a couple of taps instead of tabbing back out to Gmail.
Where this is going: Native for Legal
One more thing, and it is the one I am most excited about.
For the last few weeks we have been quietly turning Native into the daily tool for an intellectual-property law firm. Not a demo. Their real practice. And we did it on a principle we are not willing to bend: their documents never leave their building. Native reaches their records through a bridge we run on-site, so nothing sensitive ever sits on our servers. The data stays theirs, on their network, full stop.
On top of that foundation, Native learned their work. The assistant answers in the language of a trademark practice. It reads a filing and pulls out what actually matters. It shows a mark, its status, and its deadlines on purpose-built cards instead of a wall of text. And it watches the docket so a renewal or a response date never quietly slips past.
We call this Native for Legal, and it is just the beginning.
Here is the part that matters, though. We are not chasing industries for the sake of logos on a page. The whole point of Native, from day one, has been to put a genuinely capable AI into the hands of every small and mid-sized business: the ones who kept being told this technology was meant for someone bigger. Legal is simply the first place we have proven how far Native will bend to fit how a real profession actually works. The same engine will keep taking the shape of whoever picks it up next.
That is the lot for this round. As always, if something here is unclear or you want the deeper detail on any single piece, write back. I read the replies.
Gaurav
