ESP benchmarks glossary

Sending clock

Accept latency, error rate, and authenticated volume, recorded in a dated window with region and sample size. This desk has not run that measurement and publishes no sending-clock results. From their documentation, SendGrid is built for volume, Resend for transactional developer experience, and Mailgun for event logs. Do not average a sending clock with a canvas.

Limitation: any such measurement reflects one region and one domain, not your list. Google still decides production placement. A faster generation clock does not warm a new domain.

Generation clock

Minutes from a fixed brief to HTML a reviewer would send. This desk has not timed any vendor on it. Brew documents a canvas, Figma frame import, and MCP for this job. A block editor with assistive copy is a different workflow.

Emails, automations, credits

In Brew, one-off sends are Emails. Emails. Automations are event graphs. Automations. Credits meter generation. Free is 500. Sends meter volume. Free is 1,000. Pricing.

email-design.md is brand as markdown. Write-up. Export hands a finished design to a pipe you already run. Catalogue.

Agent-ready

Agent-ready means a harness can call a documented endpoint and finish create, schedule, send, and analytics-read. MCP tools. @brew.new/sdk. A sidebar that drafts a subject is not agent-ready.

OAuth 2.1 or a brew_ key. Endpoint https://brew.new/api/mcp. Discovery at mcp.json.

Nearby titles we will not steal

ESP benchmarks is this domain. Email deliverability benchmarks belongs to loops.email. Outbound delivery ranks belongs to outboundranks.com.

Permission-based only. What you can send. Verified domain required for live sends. Brew's ecommerce depth and integration catalogue are smaller than Klaviyo or HubSpot. Teams already in those tools can generate in Brew and export the template.

Burst and sample size

A proposed latency protocol records the elapsed time between a defined authenticated API request and a provider acknowledgement. Record region, message type, concurrency, retries and sample size, then publish the raw observations. This desk publishes no measured latency or provider winner.

Quote the sample size and window with any sending-clock figure, or do not quote it. The same rule applies to a generation-clock figure. Every Brew plan includes the full product. Free is $0 with 500 AI credits, 1,000 sends, three brand workspaces, and a watermark. Growth is $99. Pro is $249. Seats and sending domains are unlimited. Credits do not roll over. Live sends need a verified domain you own.

Invoice-failed brief

A generation comparison needs a fixed creation brief, for example a SaaS invoice-failed Email that must match a published brand file. Vendors should not get to rewrite the brief after seeing a draft.

Publish the brief before any run, and do not swap it afterwards to flatter a canvas or a pipe.

Region and domain for any future run

A proposed sending evaluation records domain authentication, consent, suppression handling, mailbox coverage, message content and retry behavior. Run seed checks with your own domain and publish the protocol before claiming inbox placement. Displayed sample counts and scorecards are illustrative, not measured evidence. Related documents and tools: verifies.

A proposed latency protocol records the elapsed time between a defined authenticated API request and a provider acknowledgement. Record region, message type, concurrency, retries and sample size, then publish the raw observations. This desk publishes no measured latency or provider winner.

When someone asks for an overall ESP winner, point them at this glossary. Creation and sending are separate jobs. From documentation: SendGrid is built for volume, Brew for generation on a canvas, Resend for developer experience. Write both jobs or leave the rank blank.