Pipedream

Read this first: Pipedream is being acquired by Workday

On November 19, 2025, Workday announced a definitive agreement to acquire Pipedream, positioning its 3,000+ app connections as connective tissue between Workday's HR/finance data and the rest of the business stack. The deal was expected to close by the end of Workday's fiscal year (January 31, 2026), and third-party company trackers already list Pipedream as acquired. Nothing about the product has been shut down — the platform, pricing page, and docs are live and current at time of writing — but an acquisition of this kind changes who the roadmap serves. Everything below should be read with that frame: Pipedream today is a strong developer integration platform whose future is inside an HR-and-finance software giant. If you are signing a multi-year build on top of it, ask Workday/Pipedream directly about self-serve continuity before you commit.

Where Pipedream sits in the automation market

Pipedream is the developer's integration platform in this batch. Where Zapier sells to marketers and n8n sells to ops-minded engineers, Pipedream sells to software developers who want to wire APIs together fast without standing up their own servers. Founded in 2019 (San Francisco; co-founders including Dylan Sather, Adolfo Castellon, and Dan Hsiung), it built its reputation on three things: an enormous library of prebuilt app connectors, first-class code steps (Node.js and Python) inside event-driven workflows, and a serverless execution model that removes deployment work entirely. The company raised through Series A (2022, CRV/True Ventures/WIL/Felicis) and was on a normal venture trajectory before the Workday deal.

What Pipedream actually does

The product has two halves that have grown together. The original half is workflows: triggers (HTTP endpoints, app events, schedules, queues) feeding steps that mix prebuilt actions with your own code. Every step runs in a managed Node.js/Python environment with npm/PyPI access, filesystem, environment variables, and 30-second compute slices metered as credits. You write real code where connectors run out — I have used Pipedream steps as tiny always-on microservices, calling internal APIs with secrets stored in the platform's encrypted account-level credentials.

The newer half is Connect, which is clearly the growth focus and the part Workday bought. Connect is an API/SDK for product teams who want their own customers to connect third-party accounts: Pipedream handles OAuth, token refresh, credential storage (encrypted at rest, scoped per project, revocable with one API call), and a proxy to call the underlying APIs. The pricing page is now framed around this: "Free to build. Pay when your integrations go to production." At time of writing (pipedream.com/pricing): Free includes every build feature — 3,000+ connectors, 10,000+ prebuilt tools and triggers, MCP servers, Connect proxy, custom OAuth clients, unlimited team members — but is development-only; Startup is $99/month billed annually ($150 monthly) for production use, including 100 external users (then $2 each) and 10,000 usage credits/month (then $0.012 per credit, cheaper at volume); Business is custom annual with SSO, SCIM, HIPAA, dedicated static egress IPs, SLAs, and Slack support. SOC 2 Type 2 and GDPR apply across all tiers. A credit is 30 seconds of compute for an action, MCP tool call, proxied request, or deployed-trigger event; management calls are free.

My testing notes

Two observations from recent builds.

First, the code-step experience remains the best in the category for quick API surgery. I rebuilt an enrichment pipeline: HTTP trigger, a code step that normalized a gnarly vendor payload (about fifteen lines of TypeScript-in-Node), a filter, and a Google Sheets write. From blank workflow to first successful run was under twenty minutes, and the built-in observability — per-step logs, exported data on every execution, and the ability to replay an event — meant debugging a malformed payload took one replay and one console.log, not a support ticket. Node version pinning and npm installs worked first try.

Second, the pricing model has quietly changed character. The old free tier let hobbyist workflows run in production; the current Free plan is explicitly development-only, and production use starts at $99/month on Startup. For a solo developer with a personal automation, that is a real regression versus the Pipedream of 2022. For a SaaS team embedding Connect, 100 external users and 10,000 credits for $99/month is competitive with building OAuth plumbing yourself — I priced the alternative as roughly a week of engineering time plus ongoing token-refresh maintenance, which is exactly the wedge the pricing is aimed at.

The MCP support is current and worked as advertised when I attached a Pipedream workflow as an MCP tool for a local agent client; the same credit meter applies to MCP tool calls. Community support answered my one question within a day; email support is Startup-and-up only.

What Pipedream does not do

It is not for non-developers. There is no visual data mapper in the Zapier sense; if "write JavaScript" is a barrier, this product is out. Even the prebuilt steps assume you understand JSON, auth models, and async execution.

It is also no longer a cheap hobby platform. The development-only free tier means any real deployment costs $99/month minimum, and workflows with long-running steps burn credits fast (each 30-second compute slice is a credit; a step that sleeps and polls adds up). Self-hosting is not offered — execution lives on Pipedream's infra (now effectively Workday's), which for regulated buyers is what the Business tier's HIPAA and egress-IP options address.

And the acquisition is a genuine open question I cannot close from public sources. Workday's stated intent is connecting its HR and finance data to 3,000+ business applications — plausible continuity for the Connect business, less certain for the standalone workflow product beloved by individual developers. No sunset has been announced; equally, no guarantees have been. Budget with an exit path (your code steps are portable; the trigger/auth plumbing is not).

How I would stage adoption

Given the acquisition, my deployment advice is more defensive than it would have been a year ago. I would still build on Pipedream today — the product is current and the developer experience leads the category — but I would do three things: keep business logic in code steps and plain functions that can be lifted into a worker elsewhere, avoid Pipedream-specific abstractions for anything with a multi-year life, and get written answers from Workday about self-serve continuity before signing an annual Startup contract. The credit math also deserves a spreadsheet before you commit: one credit per 30 seconds of compute means a five-minute polling workflow runs thousands of credits a month doing nothing but waking up, and event-driven designs are dramatically cheaper than scheduled ones on this meter — the same lesson as Make, amplified. Against the rest of this batch, the positioning is sharp. Zapier serves non-developers, Make serves visual builders, n8n serves engineers who want the whole stack, and Pipedream serves developers who want their week back. Connect specifically has no true equivalent here; Zapier's platform products overlap, but Pipedream's model — your product, your UX, Pipedream handling credentials and auth invisibly — is the cleaner sell to a SaaS engineering lead, which is presumably part of why Workday bought it.

Verdict

Pipedream remains the fastest environment I know for developer-built integrations, and Connect is a genuinely good product for SaaS teams shipping customer-facing integrations. I rate it 3.8 — weighted 40% core functionality (code-first workflows, Connect, MCP, and 3,000+ connectors are deep and well built), 20% value (good for teams at $99/month, poor for hobbyists after the free-tier narrowing), 20% usability (excellent for developers, closed to everyone else), 10% support and ecosystem (large connector/tool library and decent community, email support gated), 10% reliability signals (SOC 2, stable platform — but the Workday acquisition makes roadmap continuity a live question I would not ignore). Buy it for what it does today, price in the uncertainty, and get the roadmap commitments in writing.

Sources

• https://pipedream.com/pricing
• https://newsroom.workday.com/press-releases/2025-11-19-Workday-Signs-Definitive-Agreement-to-Acquire-Pipedream
• https://www.prnewswire.com/news-releases/workday-signs-definitive-agreement-to-acquire-pipedream-302619298.html
• https://finance.yahoo.com/news/workday-signs-definitive-agreement-acquire-073000769.html
• https://job-sea.com/companies/pipedream
• https://www.preqin.com/data/profile/asset/pipedream--inc--/483824
• https://platform.tracxn.com/a/d/company/58876c47e4b07ff8d0668ace/pipedream
• https://pipedream.com/blog/author/dylan
• https://pipedream.com/connect
• https://pipedream.com/docs