Why most referral programs
die in a Google Doc.

Every B2B SaaS founder we've spoken to this year knows referrals are their best channel. Referred customers spend more, stay longer, complain less. Almost none of those founders have ever shipped a program. This essay is about why.

The conversation always follows the same arc. A founder is staring at a CAC number that doubled. Someone on the team pastes a screenshot of a YC talk about word-of-mouth. A Google Doc gets created. The doc has a section called "Referral program" and three bullet points. Two weeks later, the doc is dead. Nobody shipped anything. The CAC number is still doubling.

The usual explanation is that founders are busy. That's true and it's not the whole story. Founders ship things harder than this every week. The problem isn't bandwidth — it's that the existing options for "just run a referral program" all ask you to do something unreasonable before you've proven the program will work.

The four things that kill it.

One: price before proof. Most tools in this category charge $49–$149/mo before your program has sent its first reward. That's rational pricing on their side — they have to make money too — but it collides with the founder reality that nobody wants to add a line item to Stripe for a thing they haven't proven generates revenue. The $49 is small; the decision energy it asks for is not.

Two: affiliate portals feel wrong. The referral UI lives at rewardful.com/yourbrand or firstpromoter.com/yourbrand. Your best customers — the ones loyal enough to actually refer — now have to log into someone else's product to see their earnings. At the moment they're most bought-in to your brand, you hand them someone else's. It feels rented. Because it is.

Three: payouts are a spreadsheet. Most tools at the founder-accessible price point export a CSV at the end of the month. You then open PayPal or Wise, paste rows one at a time, hit send, and pray. At scale this is a finance-team job you don't have a finance team for. At small scale it's embarrassing — you built a product that moves money automatically all day, and your own referral program moves money like it's 2014.

Four: tax compliance is on you. When a US partner crosses $600 in a calendar year, you owe them a 1099-MISC. Nothing in most of these tools tracks that threshold for you. So the Google Doc has a third bullet: "figure out 1099s later." Later never comes. The program dies in December when someone finally runs the math and realizes the company is out of compliance and the CFO is going to have a bad week.

Any one of these four would survive a scrappy founder. All four together are why the Google Doc wins.

What the Google Doc actually is.

Here's the thing people miss: the Google Doc isn't a failure to ship. The Google Doc is what you ship when the tools are bad. It's a rational response. "I will manually DM my top ten customers, offer them $250 per referral I can verify in Stripe, track it in this sheet, pay via Stripe balance at the end of the month." That works. It works until you have more than ten advocates, anyway. Then it collapses under its own weight — untracked clicks, unclear attribution, missed payouts, angry emails from partners who referred someone in February and still haven't been paid.

So the question for anyone building in this space isn't "how do we beat Rewardful?" It's "how do we beat the Google Doc?" Rewardful is the second choice. The Google Doc is the first choice and always has been.

What replaces the Google Doc.

Four things, in the order they need to be true:

Free until it works. If the tool costs zero dollars until the program generates money, the decision cost collapses. You don't need a meeting about it. You don't need budget. You drop a script tag in and see what happens. If nothing happens, you uninstall and owe nothing. The founder's question stops being "is this worth $49/mo to find out?" and becomes "why wouldn't I try this?"

Embedded inside your product. The referral widget renders in your own settings page, your own billing portal, your own post-onboarding state. No external portal. No handoff to someone else's URL at the moment of highest trust. The advocate stays inside the product they love. You stay in charge of the surface.

Payouts move themselves. When a criterion trips — referred customer pays their first invoice, refund clawback window clears — money moves. Stripe Connect fires a transfer. No spreadsheet. No PayPal paste. No monthly dread. The advocate sees their money arrive on the day you said it would, every time.

Tax is handled automatically. W-9 at partner signup, 1099 threshold tracked per-partner, year-end forms generated when someone crosses $600. This isn't glamorous — it's operationally the most important thing on the list. If your tool doesn't do this, you don't have a referral product, you have a payout simulator that will break your accounting in December.

Why this finally works in 2026.

The technical pieces have existed for a while. What changed isn't capability — it's pricing model and product surface. Stripe Connect is mature enough to automate payouts at zero marginal cost. Cookie-less first-party attribution is practical enough to run without breaking in Safari. The <script> tag is small enough to drop in any modern stack without a build step. The founder audience has enough pattern-recognition — from Vercel, Supabase, Resend — to recognize that a tool can be technical, opinionated, and free-until-it-works and still be legitimate.

Which brings us to the actual argument: the category's pricing is upside-down. The correct pricing for a referral product is "nothing until the program produces revenue, a share of the revenue once it does." Not because it's founder-friendly, though it is. Because it aligns the vendor's incentive with the customer's outcome. If the program doesn't work, both sides lose. If the program works, both sides win in proportion. That's the only pricing shape that makes sense for a tool whose job is to produce revenue.

Everything else — the embedded surface, the automated payouts, the tax compliance — is table stakes. The wedge is the pricing.

If you're the founder with the Google Doc.

Close it. Try a tool that costs nothing, installs in 5 minutes, and only bills you when your program is already making you money. If it doesn't work, you've spent an afternoon. If it does, you've unlocked the channel that you already knew was your best one — the one the Google Doc was going to get to eventually, and then never did.