Fivetran's Long Tail Vulnerability
Conor McCarter, co-founder of Prequel, on Fivetran's existential risk
Fivetran wins by being the safest choice for the biggest, most business critical data pipes, not by covering every app a company might use. Building and maintaining each connector is expensive, so the catalog naturally tilts toward sources with the most demand, the highest data volumes, and the largest budgets. That leaves a real opening in the long tail, where younger SaaS vendors still need warehouse exports to win deals, but may not be important enough for Fivetran to prioritize quickly.
-
Fivetran’s model is to hand build and monitor roughly 150 to 200 connectors, then keep them working as source APIs change. That creates premium reliability, but it also means connector roadmap decisions are shaped by where the most customer dollars and support urgency sit.
-
The long tail still matters because data teams do not just analyze Salesforce and Stripe. They also want data from niche support, marketing, finance, and vertical SaaS tools, especially as startups move upmarket and buyers begin to expect warehouse access as a normal product feature.
-
That is why alternative models emerged. Airbyte pushes connector creation toward its community to cover more sources, while Prequel helps SaaS vendors ship their own native warehouse connectors so they can keep the revenue, control the schema, and serve customers Fivetran may never rank highly enough.
This pushes the market toward a split structure. Fivetran remains strong on high trust, high volume connectors and database replication, while more of the long tail gets handled either by community maintained connectors or by software vendors building warehouse exports into their own products. Over time, native warehouse access becomes a standard feature for serious B2B SaaS sellers.