Fast: the checkout network lesson is merchant activation, not one-click speed

External-source research synthesis, not an OfferSignal client result. Based on the retained October 2026 research and source-specific dates.

Reducing checkout friction is easy to describe and difficult to make habitual. A shopper may want one stored identity, payment method and delivery profile across many stores, while each merchant still has to integrate the service and generate enough repeat demand to justify it. Fast, the San Francisco startup founded in 2019, is a useful failure case because it reached public launch, raised substantial funding and named major integrations, but permanently discontinued Fast Checkout on 15 April 2022. 1 2 3 6

What Fast built and how it worked

Fast Login was already in market in March 2020; checkout was then moving beyond a limited rollout. The first Fast-button purchase collected email, name, phone number, address and card information. Fast stored those credentials so a later Fast-enabled purchase could be completed quickly, although TechCrunch noted that changing devices required logging in again. The company said it earned a spread between what it charged for payment processing and what it paid Stripe, which was both a major investor and the underlying payment rail. 1 2

Fast’s September 2020 launch announcement described checkout for every device, platform and browser, with select merchants live and additional sellers invited to sign up. The proposed journey could bypass a conventional cart, track deliveries and support reordering. In March 2022, Fast and The Honest Company described a broader implementation connected to Honest’s ecommerce and subscription platforms: a button on product pages, links from articles, ads, email and QR codes, packaging QR codes for reorder, and a batching window intended to reduce extra shipments. This establishes an announced integration scope, not a measured conversion or retention result. The announcement came only 18 days before Fast’s service ended. 3 5 6

The reported growth did not establish durable demand

Fast raised a $20 million Series A led by Stripe in 2020 and a $102 million Series B led by Stripe and Addition in January 2021. The company told TechCrunch that checkout gross merchandise value had “more than tripled each month,” but no starting value, cohort or exact measurement period was supplied. By January 2022, Payments Dive reported Fast’s claim of availability on about 1,000 ecommerce sites and 650% company growth; neither figure provides an active or paid denominator, production-versus-pilot split, order volume or retention. 1 2 4 6

The later outcome record is more concrete but still source-qualified. TechCrunch reported 2021 revenue in the six figures and burn as high as $10 million per month. NPR reported approximately $600,000 of 2021 revenue, attributing it to The Information and former employees, and described concerns that the product was spotty and not used across all products of merchants presented as major partners. Payments Dive reported failed fundraising attempts, nearly 400 employees in a January interview and the April discontinuation. These reports show a large gap between disclosed revenue, headcount and spending; they do not prove that one decision alone caused the shutdown. 6 7 8

What the case does—and does not—apply to

For a seller of standard, already-priced products, Fast’s historical pattern illustrates direct product-page checkout, stored identity and payment, subscription connectivity and QR reorder. It is an illustrative application, not a current vendor recommendation or reported industrial deployment. Fast provides no evidence of capability matching, drawing and revision control, material or tolerance validation, equipment and capacity checks, co-design, estimate-to-binding-offer governance or procurement authority. For custom manufacturing, checkout would begin only after the harder qualification and commercial-control work was complete. 2 5

A replacement pilot should therefore separate discovery from transaction authority. Maintain canonical customer, product, revision, unit, currency, tax, material and availability records, with source timestamps and conflict handling. Measure eligible checkout sessions, completed orders, repeat buyers, net revenue, chargebacks, staff review time, integration effort and exceptions over a defined period. Require permissioned human approval for refunds, discounts, substitutions, non-standard quotes and binding orders; prevent duplicate actions and test malicious or stale instructions in retrieved content. A claimed site count or GMV growth rate is not a substitute for these denominators.

Decision and current status

Do not select Fast as a current checkout, artificial-intelligence visibility or commerce-integration vendor: no current product, pricing, support, API or maintained repository was verified, and service ended in 2022. Use the case to test merchant activation, repeat demand, integration depth and unit economics before scaling a successor. Public sources reviewed disclose no Fast subscription, transaction, setup or integration price. Historical funding references also require care: the retained record supports the $20 million Series A and $102 million Series B, while public reports cite approximately $124 million or $124.5 million cumulative funding with differing rounding and source methods. 1 2 4 6 7

Sources and next reading

Explore OfferSignal · Focused Business & AI Workshop · All research

References