Company profile · 4 min read
Lopus: A governed semantic layer for revenue decisions
Lopus unifies business data, captures metric definitions and provides no-SQL analytics, alerts and root-cause investigation for RevOps and BizOps teams.
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
Published · Updated
Lopus is a semantic layer and business-analytics platform for RevOps and BizOps teams that need answers grounded in their own definitions. It fits a company whose CRM, billing, product and support data are connected badly enough that every important question becomes a data-team ticket.
What it does
Lopus’s homepage describes a layer that unifies CRM, revenue and customer data, maps business logic and lets teams ask questions without writing SQL or waiting for dashboards. It also advertises alerts and root-cause investigations when a metric moves. The core product decision is to define a metric once—along with its edge cases—and reuse that meaning across answers, reports and investigations.
The security and governance surface is part of the pitch. Lopus says each instance and its data are isolated, access can be permissioned by dataset and business logic, and customer data is not used for training. Those are company claims to verify against the deployment and contract. A semantic layer is only as useful as the definitions and source joins it can defend.
Pricing is public. The pricing page lists a Growth plan at $1,999 per month with unlimited seats, more than 500 connections, 10M rows per month, six-hour refresh and root-cause analysis. Enterprise is custom, with hourly refresh, SSO and self-hosting listed. The same page repeats a contradictory “Free Plan” label next to the $1,999 price, so procurement should confirm the current plan names and limits before relying on the table.
Founder context and tradeoffs
The YC profile identifies Aamish Ahmad Beg and Danylo Borodchuk as founders. Aamish’s public background includes Dartmouth computer science, LLM-centered research and DARPA’s DIGIHEALS project; Danylo’s includes technical R&D at Dartmouth’s DALI Lab and product and systems work. That context supports the semantic-layer thesis without proving that Lopus will resolve a buyer’s data cleanly on day one.
The main tradeoff is implementation depth versus speed. Lopus promises answers in minutes and no engineering tickets, but the buyer still has to decide which systems are authoritative, who can change metric definitions and how errors are reviewed. A fast wrong answer is more dangerous than a slow dashboard when it drives revenue decisions.
Editorial take
I would shortlist Lopus for a growing SaaS company with recurring RevOps questions, a small data team and enough source-system mess to justify a semantic layer. The $1,999 starting point makes the pain threshold explicit. I would start with one revenue metric and one root-cause workflow, then test whether the definitions survive real operator questions.
Quick facts
| Field | Sourced detail |
|---|---|
| Product | Semantic layer, business analytics, alerts and root-cause investigation |
| Buyers | RevOps, BizOps and growth teams |
| Public pricing | Growth $1,999/month; Enterprise custom; public plan labels need confirmation |
| Public limits | 10M rows/month, 500-plus connections and six-hour refresh on Growth, as listed |
| Main gate | Source joins, metric governance, permissions and definition review |
Sources checked
| Source | Checked |
|---|---|
| YC profile | 2026-09-19 |
| Lopus homepage | 2026-09-19 |
| Lopus pricing | 2026-09-19 |
| Lopus blogs | 2026-09-19 |
Cohort context
Lopus is listed in Winter 2025. In our 2026-09-18 directory snapshot, 104 of 165 listed companies in that cohort have YC’s primary industry label B2B (63.0%). This is a current-directory comparison, not an original intake count or a performance ranking. Nine-cohort dataset.
Public website snapshot
Observed 2026-09-19T16:19:37.899Z in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.
| Signal | Homepage observation |
|---|---|
| Product description metadata | Observed |
| Canonical link | Observed |
| H1 or H2 heading | Not observed in this response |
| Typed structured data | Not observed in this response |
| Docs/developer link | Not observed in this response |
| Pricing link | Observed |
| llms.txt link | Not observed in this response |
| Markdown alternate | Not observed in this response |
Public observations · Collection method. Missing links here do not establish that a capability or file is absent elsewhere.
About the author
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.
First1000 ↗ · X ↗