Back

Playbook

The SaaS Integration Outreach Message That Gets a 32% Reply Rate

August 28, 20268 min read

Short answer: a good fit doesn't get you a reply — a good message does. Before you write to anyone, run the candidate through a fit filter and a quality filter (does the moment even need a partner, or could you just build it). Then send a six-move message that states the overlap in their language and asks two qualifying questions before either of you spends real time. The last time I ran this exact sequence it converted at 32%.

Why most integration attempts die at outreach, not at fit

Integrations are the most underrated distribution channel in SaaS right now, for a simple reason: customer acquisition costs keep climbing because everyone is fighting for the same attention on the same three platforms. An integration sidesteps that. You're not buying attention — you're borrowing an audience that already exists, from someone who benefits when you do.

The frameworks for finding a fitting partner are well covered — I've written up the four-gate version and the ICP × JTBD × value-chain framework it narrows from, and I'm not going to repeat either here. What those pieces don't cover is what happens after you've found the fit: how to build the shortlist without a directory, how to tell a good candidate from a time-waster before you reach out, and — the part that actually decides whether anything gets built — what to put in the first message. That's what this is.

Revenue share, so nobody carries the risk

One framing decision up front, because it's what makes the rest of this work: if it costs nothing to acquire these users, the deal cannot be that your partner pays upfront. It has to be revenue share. Prove first that the users want it, then pay. If it doesn't work, nobody's out anything and you move on. If it does, you both take a slice of an audience you never had to market to.

For reference, the collaborator's share in these deals typically lands between 20% and 30%. Sometimes more, sometimes less — that part is yours to negotiate.

The moat filter — run this before you write a single message

A candidate can pass every fit gate and still be a bad use of your outreach budget, because fit tells you the partnership makes sense, not that it's worth doing instead of building the feature yourself. Before reaching out, check whether the candidate has a moat. If they have none, you can simply build what they do, and there's no reason to put someone else's logo — and someone else's revenue cut — into your product.

The three worth checking (covered in more depth in the three defenses that survive when intelligence is free):

  • Network effect. Does the product get more valuable with every new user that joins? If no, they don't have one. This is my favourite of the three, because it's mostly a design decision rather than a bill — it comes down to getting over the hump of critical mass.
  • Proprietary data. Do they hold data nobody else can access? This matters most if either of you wants to train an AI on it — usage is what makes data good and hard to copy, and usage is expensive to buy.
  • IP and licences. Self-explanatory, and there's no way around it: you pay up. Think Stripe, whose moat is the licences required to move money.

A worked rejection: I once looked at a tool that rewrites AI copy to sound more human. Network effect? No. Data? Maybe, but who's qualified to judge whether a sentence sounds robotic? IP? Only if they'd patented the system prompt, which they hadn't. So what's stopping me from building that myself? Nothing. Skip.

One more test worth applying: if a candidate's whole pitch is "grow faster on X" and the founder has 3,200 followers, something is off. If they had 100K followers built with their own product, they'd be their own proof and I'd be sold.

Where the shortlist comes from

Start with the people you're already connected to, not a cold market scan. I have a genuinely small X account — 238 followers. I clicked through every single one and read what they were building. It took twenty or thirty minutes.

Out of 238 I shortlisted 17. That's 7%.

If you're a SaaS founder who's been around on X for a while, expect a similar number, maybe better depending on how niche you are. The point is that a real chunk of the people already following you are plausible collaborators, and you've never looked.

The six-move message

This is not a sales message. You're initiating a partnership, and you want to find out whether there's a real overlap before either of you spends real time. Mine follows the same six moves every time:

  1. The honest opener. "Hey man, I saw that we've been connected here for a while and I haven't reached out yet."
  2. Something specific you noticed. "I took a quick look at [product] — I love the website, it's really unique. The mobile version isn't as good as the desktop one, though." Specific, and not flattery.
  3. Why you're actually here. "The reason I'm reaching out is that I'm currently helping an ERP platform find integration opportunities with other SaaS apps. It's been around about 15 years and has 200 paying customers."
  4. The overlap, in their language. "They're looking for integrations that help their customers with customer-relations problems and conversion — just like [product] does."
  5. Two qualifying questions. "How far along are you in the SaaS journey? Do you have revenue? Users?" You need this to know whether it's worth going further, and asking directly is faster than guessing.
  6. The offer, and out. "If you're interested I'd love to connect you two and see if you can help each other grow. Have a fantastic week ahead."

Ask the qualifying questions — don't skip them because they feel presumptuous. I once sat down on a call with a founder assuming we overlapped, and only on the call did I learn what his product actually did — it was something else entirely. Sorry for wasting your time; shit happens; move on. Two questions in a DM would have caught it.

What happened when I sent it

The last time I ran this exact sequence — moat filter, then the six-move message, sent to a shortlist built from my own network — it converted at 32%. Not 32% signed integrations; 32% of the people I messaged wanted to keep talking. That's the number that matters at this stage, because a reply is where you find out whether the fit you assessed on paper survives contact with the actual founder.

The whole thing costs you a prompt and an afternoon of reading. That's the reason it's the cheapest customer acquisition available right now — not because integrations are new, but because the two things that made them expensive (finding the right partners, building each integration by hand) have both collapsed toward zero in the last two years.

What to do this week

  1. Open your followers list and read it. All of it. At 238 followers it took me half an hour.
  2. For each plausible one, run the fit check — same ICP, same job to be done, a different step in the value chain. (Full framework: here.)
  3. Of those, keep the ones with at least one moat: network effect, proprietary data, or IP and licences. If a candidate has none, build the feature yourself instead.
  4. Send the six-move message. Include the two qualifying questions.
  5. When they reply: chat, get on a call, scope what the integration would look like, build it, test it, and set up the revenue share at 20–30%.

Related reading:

Ordana does this matching for you — it finds SaaS collaborators with the same customers and a different place in the value chain, and handles the revenue share. joinordana.com.