Brew - Like Claude design for email marketing
by•
Brew is the fastest way to design and send beautiful, on-brand emails and automations that render perfectly in every inbox. Describe a campaign or a multi-step automation in plain English, and Brew builds the whole thing in seconds: copy, design, audience, and logic. Works with any AI agent: paste our docs into OpenClaw, Viktor, Claude, or Lovable. No lock-in: send from Brew or export to your ESP. Free to get started.
Replies
The Anthropic example sold me more than the video did. Most brand extractors grab the hex codes and call it done. Appreciate the integrations you offer, especially the one with figma.
Brew
@daria_goroshko1, this made my day. "Most brand extractors grab the hex codes and call it done" is exactly the cliff we did not want to fall off. A brand is fonts, voice, imagery, spacing, the whole feel of how a company shows up. Anything less and your email reads like someone described your brand to an intern over the phone. Glad the Anthropic example landed, that one was a deliberate test of "can we actually reproduce a brand we did not build," and the Figma integration was where it really started to click. Thanks for paying close attention. ❤️🔥
Cool concept. Just one question from a guy that does a bit of lifecycle mailing himself. You say that it renders perfectly in every inbox, but does it change anything about the actual placement? I honestly struggle more with my welcome email landing in the Promotions tab in Gmail, than how it looks. Is that something Brew helps with, or does that rely completely on the domain I send from and my setup?
Brew
Genuinely great question @firecalculatorhq. Honest answer: placement is mostly determined by factors outside the email tool. Sender reputation, authentication (SPF, DKIM, DMARC), list quality, and engagement history matter more than the HTML.
What Brew does: generates clean, conservative HTML that does not trip spam filters, and handles SPF/DKIM/DMARC on your sending domain when you send through Brew. For batch sends we recommend warming the domain gradually. What we cannot do is guarantee Primary tab placement, since Gmail and Outlook decide that based on recipient behavior and sender history. Any tool that promises 100 percent Primary is overselling.
Short version: Brew removes the technical reasons your email might trip filters, but the welcome-to-Promotions problem is mostly a sender-reputation game, and that one is yours to win or lose.
@philip_sorensen Thanku for an honest answer! You could've easily said "yeah, Brew fixes that" and sold it to me, but you actually owned the fact that placement is your own responsablilty. That was actually what I needed to hear.
I've stared myself blind regarding how the email looks, when the problem probably sits in my domain and engagement history.
One last question while I've got you here: for a little list that I've barely begun building, does it matter wether i set DMARC to p=none or quarantine from start, or is that just overthinking at this stage?
Brew
@firecalculatorhq, happy to give you a straight answer here.
Short version: start with p=none, move to p=quarantine after two weeks of clean reports. p=none lets receiving servers report on your behalf without acting, which gives you a window to see if anything legitimate is failing SPF or DKIM alignment before it starts getting blocked.
With a small list you cannot afford to lose deliverability while you are still figuring out setup. Once reports come back clean for a couple of weeks, move to quarantine. p=reject is the destination but only worth it once you are sure nothing legitimate is getting caught.
One thing to add: SPF and DKIM need to be set up correctly first. DMARC only enforces what those two say.
Rooting for you with the list build.
That was exactly what i needed! Thanku for taking the time to explain the order. p=none first, then quarantine when the reports are clean, and get SPF/DKIM right before i even touch DMARC. I'll set it up like that now. And thanku for your cheering, that's well needed now :)
@Migma AI @Dreamlit AI got some real competition here. :D Congrats on the launch Brew team!!
Brew
@himani_sah1 Thanks Himani! We're big fans of Migma and Dreamlit :) There's many talented teams working in this space and for a good reason!
Brew
@himani_sah1 we have massive respect for the other companies in this space as @thomas_park2 said. At the end of the day, we competing against the big guys - MailChimp, HubSpot, Braze, etc. - not so much other start-ups. We're all rooting for each other ❤️🔥
How complex can the automations get? Wondering if it handles conditional branches based on whether someone opened or clicked.
Brew
@rajmishra10 Hey Raj! Automations can get pretty complex in Brew. We want to abstract away as much of the manual process by enabling you to simply ask for the type of automation you want and the agent will create the flow. Currently the automations are triggered via an API call to Brew or a webhook event from a third party service like Clerk or Supabase. We have click / open logic gates on the roadmap but will support it very soon!
The "describe a campaign in plain English, get the full thing rendered" framing is exactly what email needed — most teams aren't bottlenecked on having something to send, they're bottlenecked on the production tax of every send. I work on StoryRoute on the travel side and we see the same dynamic: people will narrate what they want a city walk to feel like, but they won't sit down and design the route step-by-step. The product that closes that gap wins. Congrats on the launch.
Brew
Thank you @samir_asadov , you put it better than we do. "Production tax" is exactly the phrase. The bottleneck was never the idea, it was the eight steps between the idea and the send, and that is where everything quietly dies. Love the StoryRoute parallel too. Narrating the feeling of a walk versus plotting every turn is the same gap, and you are right that whoever closes it wins.
Best of luck with it, and thanks for the kind words. ❤️🔥
@philip_sorensen The eight-steps observation is the part I always come back to. The unit cost of any one step looks trivial ("it's just two minutes to drop in the right image"), so it never gets prioritized — but compounded across every campaign and every team member, those two-minute taxes are where the actual marketing budget goes, just in unbilled hours nobody can see. The tools that win this space are the ones that collapse the eight to two, not the ones that polish step seven. Looks like that's exactly the call you made.
Brew
Exactly right @samir_asadov , and "unbilled hours nobody can see" is the cleanest way I have heard it put. That invisible cost is precisely why the category stagnated. Everyone optimized step seven because each individual step looked too cheap to bother rethinking, so nobody questioned the whole chain. Collapsing eight steps to two was the entire bet. It is a harder thing to build than a better step seven, but it is the only version that actually changes the economics for a team. You clearly think about this the same way we do. If StoryRoute is collapsing that same gap on the travel side, you are going to do well.
Askmeety
Brew
Thank you @nalin_rajendran, and you have put your finger on the exact thing we worry about most...
Generic, bot-ish email is worse than no email. The way we fight it is by capturing both the art and the science of your brand: not just colors and fonts, but your actual voice and the feel of how you communicate, so the output sounds like you rather than like a model. On top of that we make it easy to experiment with different styles for your brand, so you are never locked into one look. Here is an example of what that looks like for one of our users.
"Renders perfectly in every inbox" is the claim I'd stress-test first, because Outlook's rendering engine has buried prettier email tools than this. When Brew generates a layout, is it producing battle-tested table-based HTML under the hood, or modern CSS that looks great in Apple Mail and falls apart in Outlook 2016? That one answer decides whether I'd trust it for a real send.
Brew
Great question to stress-test first @sounak_bhattacharya, and the honest answer is that the safety lives in how generation is constrained, not in a cleanup pass afterward. We do not let the model emit free-form modern CSS and hope it survives Outlook. Output is locked to a vetted set of email-safe building blocks, the battle-tested kind that hold up in the Word rendering engine, so it cannot produce something that looks great in Apple Mail and collapses in Outlook 2016. On top of that we have done a lot of testing and evals across real clients to find what actually holds everywhere. Outlook specifically is the one we obsess over. Happy to go deeper if useful.
BetterClaw
renders correctly in outlook is doing the real heavy lifting here. you can build the most beautiful email in the world and outlook will find a way to break it. if this actually solves the rendering nightmare across inboxes thats worth it on its own
Brew
You are speaking our language @tina_chhabra. You can build the most beautiful email in the world and Outlook will still find a way to break it, which is exactly why we built for it first instead of treating it as an afterthought. Getting the rendering right across inboxes was the whole point, and honestly the part we are proudest of. Thanks Tina.
The "renders perfectly in every inbox" claim is what caught my eye —
Outlook's rendering quirks alone have killed so many email tools.
How are you handling that under the hood?
Brew
Outlook is exactly the one that has killed prettier tools than ours @hirogure, so we treated it as the hard problem from day one. Without giving away the recipe, it comes down to a combination of things: generation constrained to email-safe markup rather than free-form CSS, plus a lot of testing and evals across real clients to make sure it holds up. The result is rendering you do not have to babysit, Outlook included.
Gents, this is an amazing tool but am I going crazy or isn't there any undo button? I've been trying to perfect an e-mail for almost an hour but always seem to mess things up by dragging or fiddling after which my only option is to start over with a previous version but losing all the steps in between that said version and the point where I messed up. What am I missing? *sweating profusely : )
Brew
@kristof_bogaerts1 Hey Kristof! Great question. We have control + z or command + z for manual edit undos and versioning for any agent chat changes. Let me know if you have any issues with it, happy to walk you through it in a personal loom or call as well :)