Explores & dashboards
Explores are powerful but every person who opens one is a per-seat line item — and the dashboards themselves are slow enough on large data that PDTs are the de-facto answer.
What people actually say
✕Per-user licences on top of the platform fee are Viewer ~$400/year, Standard ~$799/year and Developer ~$1,665/year — a 50-person team with 40 viewers, 8 analysts and 2 developers lands around $16,000 in user licences alone before the platform fee.
Source: Toucan Toco, Looker Pricing 2026 ↗✕Performance complaints are consistent in 2026 reviews: slow dashboard load times, broken visualisations after small upstream changes, and bugginess on large datasets — reviewers report quota limits surfacing as errors mid-meeting.
Source: Google Developer Forums, Looker Studio reliability ↗✕Persistent Derived Tables — the cache layer that makes Explores tolerable on large warehouses — push query volume back through your BigQuery account, where third-party estimates put annual BigQuery costs at $50,000-$200,000 driven by Looker activity.
Source: Mammoth, Looker Pricing 2026 ↗
Dashboards are React components on Next.js or an Evidence/Metabase deployment in your own infra, querying your warehouse directly through the dbt semantic layer. No Viewer SKU, no per-seat multiplier on read-only dashboards, and the marginal cost of one more person opening a chart is one more HTTP request — not a £400-a-year line item.