The build-out has a footprint too. Here's ours.
Watermark exists to put a compute build-out on the public record. Running the platform is itself compute — servers, storage, and a working set of AI models that read documents so you don't have to. We hold ourselves to the same standard we ask of the sites we track: the numbers below are published as a lead, not a verdict, and every one is labeled with how it was derived.
Built to do less, by default
Most of what cuts our footprint isn't a program — it's an architecture decision made once and inherited by every page after. In order of how much they save:
A record's structured read is computed once, at corroboration time — not re-derived every time a page is opened. Every later visit is a cache hit, not a new inference.
Pages are mostly static HTML built from the corroborated record. No client bundles re-fetching and re-rendering state on every load.
Extraction and classification run on small, task-tuned models. Large frontier models are reserved for the handful of steps that need judgment.
Document ingestion runs on a schedule against off-peak grid windows, in batches — not triggered per visitor request.
No analytics scripts, ad pixels, or session replay. Less fetched and less run, on every single page view.
Sites still in early tracking phases sit in cheaper, lower-power storage tiers until they're active investigations.
Where models touch the record
Models are scoped narrowly — to reading, not to deciding. We don't run recommendation engines, engagement scoring, or ad targeting. Four jobs, each sized to the smallest model that does it well:
Turns a scanned permit or filing into typed fields — dates, parties, figures — the record can display and cite.
Surfaces candidate sources that might confirm or contradict an open lead. Never confirms one on its own.
Answers a question against the corpus and returns it with citations back to source pages.
Turns dense source excerpts into a readable summary for a profile or timeline entry, held for editorial review.
Compute, AI, electricity & water
Hosting & serving the record still costs more than reading it with a model — the site is mostly static pages, not live inference.
- Structured extraction 767
- Search & Ask 418
- Corroboration assist 140
- Drafting summaries 70
Structured extraction — turning a filing into typed fields — is most of the volume; drafting is the smallest and most closely reviewed.
- Grid (regional mix) · 0.0482 MWh · 90.1%
- Matched renewable (RECs) · 0.0053 MWh · 9.9%
- Direct — on-site cooling · 25.4 gal · 48.7%
- Indirect — grid generation upstream · 26.8 gal · 51.3%
AWS Cost Explorer instance-hours (by service, folded into platform functions) converted to vCPU-hours by instance size, plus GitHub Actions minutes converted by runner core count. Model inference runs on third-party/serverless infra (Anthropic, Bedrock) and is reported as AI volume, not our vCPU-hours; we run no GPU instances.
vCPU-hours x 7 W per allocated vCPU (an average-operational assumption folding utilization) x PUE 1.2. Derived ~0.0145 MTCO2e (electricity x 596.326 lb/MWh, SRVC) vs AWS's location-based estimate 0.045 MTCO2e — ~32% of it. The two agree in order of magnitude; they differ because the vCPU model captures only active instance compute, whereas AWS's estimate spans every service (storage, transfer, managed-service and idle-capacity overhead).
Modeled from electricity using published Water Usage Effectiveness (WUE) benchmarks — a site WUE for direct on-site cooling and a source WUE for the water withdrawn upstream to generate the power drawn. The two bases are reported separately, never summed across bases.
The Anthropic Admin API exposes token aggregates only, so the call count is total tokens ÷ 4000 avg tokens/call (a stated assumption), and the by-task split is modeled until the per-task workspace keys (#1080) are named so the by-workspace usage can be labeled and metered by task.
sources · AWS Cost Explorer + Sustainability/CCFT (Jul 2025-Jun 2026) · GitHub Actions/storage billing (Jul 2025-Jun 2026) · Anthropic Admin usage + cost (Jul 2025-Jun 2026) · EPA eGRID2023 subregion SRVC factors · WUE benchmarks (EPRI 2024 / Uptime Institute 2023 / NREL 2012)
Same rule as everywhere else on Watermark: every figure links to a correction.
maintained by the record team · data catalog · about