privacy

Zapier and Make: What Happens to Your Files in No-Code Tools

Zapier and Make process your personal files as they pass through. Here's what they access, what they retain, and how to reduce your exposure.

No-code automation tools have become genuinely useful for individuals managing their digital lives. Zapier, Make (formerly Integromat), n8n, and IFTTT let you connect services and automate tasks without writing code. Common personal automations include saving email attachments to cloud storage, automatically organizing photos, backing up form submissions, and archiving social media posts.

These tools work by connecting to your accounts via OAuth, reading data from one service, and writing it to another. This means they necessarily touch your personal files as part of doing their job.

Most people set up an automation, let it run, and never think about what happens to their data inside the automation platform itself. That gap in understanding is worth closing.

What OAuth Access Actually Grants

When you connect a service like Google Drive or Dropbox to Zapier, you authenticate via OAuth. This process grants Zapier a token that allows it to access your files within the scope you authorize.

The scope matters. When Zapier asks for access to Google Drive, the default scope is typically read and write access to all your files — not just the ones involved in your automation. This is a technical requirement of how the integration works, but it means Zapier has a token that could, in principle, read any file in your Google Drive.

Zapier states in its privacy policy that it accesses and processes only the data necessary to run your Zaps. This is a privacy-by-design commitment, not a technical restriction. The technical capability is broader.

To review and manage OAuth permissions you have granted, go to:

  • Google: myaccount.google.com → Security → Third-party apps with account access
  • Dropbox: dropbox.com → Account → Connected Apps
  • OneDrive: account.live.com → Your info → Privacy → Manage your privacy → Apps and services
  • Box: app.box.com → Account Settings → Apps

Review this list periodically. You may find automations you set up years ago still have active OAuth tokens.

What Automation Platforms Retain

Understanding data retention is key to evaluating the privacy profile of these tools.

Zapier

Zapier processes data in memory as Zaps execute. File content that passes through a Zap is not permanently stored by Zapier on its servers in its full form. However, Zapier retains task history — a log of what each Zap did, including metadata about the trigger event and action taken.

For some integrations, task history includes previews or summaries of the content processed. If you have an automation that emails you a summary of files processed, that content may appear in task history logs.

Zapier retains task history for a period that varies by plan tier. Free accounts may have shorter retention; paid accounts typically have longer retention. This history is accessible to Zapier administrators and, per their policy, to law enforcement with appropriate legal process.

Zapier experienced a data breach in 2022. An attacker accessed a subset of Zapier source code and, separately, task history data for some accounts. The scope was limited, but it confirmed that task history data is a real attack surface.

Make (formerly Integromat)

Make operates similarly. Data passes through Make’s infrastructure during automation execution. Make retains execution history, which may include data payloads depending on the modules used and how scenarios are configured.

Make’s privacy documentation states that data payloads transmitted through scenarios are not stored permanently after execution completes, but execution history and logs are retained. The precise retention period depends on the plan and scenario configuration.

n8n (self-hosted option)

n8n is open-source workflow automation software that can be self-hosted. This means you run the automation infrastructure on your own server or on a cloud server you control. No data passes through n8n’s corporate infrastructure — everything runs within your own environment.

For automations involving personal files, self-hosted n8n eliminates the third-party data retention concern entirely. The trade-off is that self-hosting requires technical ability (running a server, managing updates, handling SSL) that many individuals do not have.

n8n also offers a cloud version, which operates similarly to Zapier and Make in terms of data handling.

IFTTT

IFTTT (If This Then That) has a simpler automation model than Zapier or Make, and it handles smaller data payloads. IFTTT retains run history and stores connection tokens. Their privacy policy is less detailed than Zapier’s regarding data processing, which is itself informative — less transparency typically means fewer commitments.

For simple personal automations (turn off smart lights when you leave home, post to multiple social media simultaneously), IFTTT’s data exposure is limited. For automations that involve document or photo content, the opacity is a concern.

The Specific Risk with Personal File Automations

The most common privacy-sensitive personal automations are:

Email attachment backups: automatically saving all email attachments to a Google Drive or Dropbox folder. Your email may contain medical bills, legal documents, financial statements, and confidential correspondence. Running these through an automation platform means that content passes through a third-party service and potentially appears in execution history logs.

Photo organization automations: automatically organizing photos by date, location, or event. Photos moving through an automation platform expose both the image content and embedded metadata (EXIF data including GPS coordinates, camera model, timestamps).

Document processing: extracting text from PDFs or images, categorizing files, creating summaries. If you automate processing of personal documents, the content is processed by the automation platform and any AI services you connect to (OpenAI, Google AI, etc.).

Health or personal data aggregation: automations that pull from health apps, fitness trackers, or personal finance apps aggregate sensitive data and process it through a third-party platform.

The risk is not that these platforms will misuse your data — it is that they become an additional attack surface. A breach at Zapier, Make, or any connected service exposes whatever has passed through those systems.

How to Reduce Your Exposure

Scope your OAuth permissions precisely

Some integrations offer narrow scopes. When connecting to Google Drive, check whether the integration allows you to grant access only to a specific folder rather than all files. Zapier supports folder-scoped Drive connections for some automations. This limits the blast radius if a credential is compromised.

Audit your connected apps quarterly

Revoke any OAuth connection you no longer actively use. Old, unused tokens are a risk — the connection exists and the token remains valid even if you haven’t run the automation in months.

Avoid automating the most sensitive data categories

Not everything should flow through an automation platform. Medical records, legal documents, financial statements, and intimate photos benefit from being handled manually and stored in a controlled environment. The convenience of automation is not always worth the added exposure for the most sensitive files.

Keep sensitive originals in storage you control

Automations can create or move copies of files, but the authoritative version of your most important files should live in storage you have direct control over — not dependent on an automation platform’s execution to maintain.

Self-host if the data is sensitive enough

If you regularly automate workflows involving personal files and privacy is a serious concern, running a self-hosted n8n instance eliminates the third-party intermediary. The technical overhead is significant, but for professionals handling client data or individuals with elevated privacy needs, it is a real option.

The Broader Problem: Data Sprawl

No-code automation tools are a symptom of a larger phenomenon: your personal data is now scattered across dozens of interconnected services, each of which has access credentials to the others.

When you connect Zapier to Google Drive, Dropbox, Gmail, Slack, Airtable, and five other services, you have created a web where each platform has some access to your data through its OAuth connection, and Zapier has access to all of it through its centralized position.

This is valuable for automation. It is also a sprawling attack surface. One compromised credential — Zapier’s or any of the connected services’ — potentially exposes data from multiple connected platforms.

The discipline of periodic access audits, narrow permission scoping, and deliberate decisions about which data flows through automated channels directly reduces this risk.

Automation Platforms vs. Dedicated Private Storage

There is a conceptual distinction worth making. Automation platforms are intermediaries — they move data between places and process it in transit. They are not designed to be the primary secure home for your personal files.

Dedicated private storage — whether a self-hosted NAS, an encrypted cloud storage service, or a privacy-focused app — is designed with the assumption that your files stay there, protected, and are not flowing through processing pipelines.

For automations that involve personal data: be thoughtful about what you route through these platforms, audit your connections regularly, and keep the authoritative versions of your most sensitive files in a storage environment that is not defined by its ability to integrate with everything else.

The convenience of no-code automation is real. So is the data exposure it creates. The goal is to use it deliberately, not by default.

Your memories deserve better than an ad platform.

Try daftei free →
← All posts