We started Clycyo because the analytics layer of the modern stack is broken.

Every developer knows the feeling. You ship a side project. You drop in Google Analytics because it's free and everyone uses it. Six months later, you're paying a Lighthouse penalty, dealing with cookie banner code, watching your data get sampled, and trying to make sense of a UI built for a different decade.

You consider Plausible. It's clean, it's simple, it respects your users — but you also need to track revenue when a Stripe payment lands. You consider DataFast. It tracks revenue elegantly — but it sets cookies, which means a banner, which means lost conversions, which means a regulator could fine you.

So you patch together three tools. You write glue code. You re-instrument when you scale. You re-instrument again when you sell to enterprise. Each layer of glue is a chance for the data to lie to you.

The bet

We think the analytics layer should be one tool that grows with you — from your first pageview to your first $10k MRR to your enterprise SOC 2 audit. One mental model. One SDK shape. One dashboard that scales.

And we think it should respect three things at once that the industry insists on trading off:

• Privacy — by construction, not by claim
• Revenue attribution — server-side, refund-aware, multi-touch
• Developer experience — typed SDKs, real docs, open source

And it should be free at the volumes most projects live at, forever.

What we won't do

We won't add cookies “just for accuracy.” We won't bolt on session replay we can't make private. We won't charge per seat. We won't sample. We won't fabricate testimonials or vanity metrics. We won't hide the bad parts; we'll publish our weaknesses on every comparison page.

When we ship, we'll dogfood. When we're wrong, we'll say so on the changelog and the postmortem. When we win, we'll show the math.

Where we're going

The next wave of SaaS founders is shipping AI-native apps in browsers, on edge runtimes, with payment models we're still figuring out. They need analytics that handles tokens and latency and prompt-level cost without a custom event for each. They need it to ship the same way they ship everything: npm install, env var, three lines of provider, deploy.

That's what we're building. If that's you, we'd like to build it with you.