How Eight Sleep Turned Black Friday QA From All-Nighters to Automated Confidence

Midnight war rooms, spreadsheets, and anxiety

For every major sale, Eight Sleep’s QA process revolved around a two-person product team and one giant spreadsheet.

Alanah, who leads e-commerce product, would block off late nights and lock herself in a room with:

• A massive spreadsheet with many tabs
• 15+ regions, currencies, and local rules
• Variants and sizes of Eight Sleep’s flagship product, the Pods
• Hundreds of partner landing pages

They had some automated tests focused on core flows, but everything qualitative and visual was still manual: regional discounts, partner pages, language, presentation, and all the things customers actually see.

“There was always this underlying anxiety that something would break. To the best of my ability everything was covered, but you always worry there is an edge case you did not think about.”

Why Eight Sleep chose Spur

Before Spur, Alanah tried to evaluate more traditional QA tooling like Playwright-style setups. Those tools were good at asserting that “event A leads to event B,” but not at answering questions like:

• Does this discount look correct in this region for this specific product variant?
• Does the landing page that a podcast partner sends traffic to match what they are promising?
• Does this page feel like a premium brand experience to a real customer?

Most solutions still required heavy engineering time and could not replace the human-style QA Eight Sleep needed.

Spur was different for three reasons:

1. Agentic, vision-first QA
   Spur behaves like a user and looks at the site like a human, rather than only asserting events.
2. Coverage for all the messy reality of e-commerce
   Multiple regions, currencies, product variants, and hundreds of partner pages are exactly where traditional tests fall apart. Spur’s agents can exhaustively explore and compare these states.
3. No engineering lift to get started
   Onboarding could be run by product and e-commerce, with engineers only pulled in to fix issues and learn how to debug from the Spur reports.

“The other tools I looked at did not feel like they could replace human QA. With Spur, I realized it was actually possible to solve those problems.”

Rolling out Spur before Black Friday

Eight Sleep onboarded Spur in roughly two weeks, right before the Black Friday period.

1. Turning spreadsheets into a knowledge base

All of the manual context that lived in Alanah's spreadsheets and brain moved into Spur:

• Regions and their associated rules
• Types of discounts and sale configurations
• Personas and entry points:
  • Existing members
  • First-time visitors
  • Visitors coming in from partner pages

Those scenarios became the foundation for Spur tests.

2. Building agentic test suites across staging and production

Eight Sleep and Spur worked together to:

• Create suites that covered critical sale flows across regions and product variants
• Map scenarios to personas and entry points
• Run tests first on staging, then on production after launch

By the time feature flags flipped, Eight Sleep already had high confidence from staging results. Once the sale went live, they used Spur to triple-check across live environments rather than starting QA from scratch.

“It now feels like running the tests is just part of the process. I already have high confidence from staging before we even set things live.”

3. Onboarding without engineering bottlenecks

Spur’s team was hands-on, helping set up tests even before Eight Sleep fully logged in and staying responsive in Slack whenever issues popped up.

• Product and e-commerce owned setup
• Engineering only joined to fix issues and learn how to debug using Spur’s reports
• No custom testing framework or code-heavy setup required

“We were onboarded in about a week and a half. It was very quick, with no time required from engineering. That is huge for us.”

How Eight Sleep uses Spur today

Eight Sleep now uses Spur to cover:

• Critical e-commerce flows across 30+ regions
  Pricing, discounts, and product variants per region
• Hundreds of partner landing pages
  Ensuring that every audience coming from podcasts, newsletters, and other partnerships lands on a page that is on brand and error free
• Staging and production environments
  Running tests in staging for high confidence before launch, then validating again on production as a final safety net

Spur runs on a cadence so that even during code freezes, unexpected changes from third parties are still caught quickly.