Twitter API with Zapier: How to Pull X Data (2026)

By Sorsa Editorial

Updated: June 17, 2026. Reflects the current bring-your-own-key status of Zapier's X integration after the March 2026 on-off cycle, and the official X API pay-per-use rates in effect after the April 20, 2026 change.

That second path is where most read-only workflows belong, and it removes the two biggest friction points of the native integration. Sorsa API, an alternative Twitter/X API provider, gives you read access to public X data through a single key passed in an ApiKey header: no X Developer App, no OAuth screens, no approval queue. Billing is flat, so one call counts as one request no matter how many tweets or profiles come back, on plans that start at $49 a month for 10,000 requests. Inside Zapier you call it from a Webhooks by Zapier or API by Zapier step, aim it at an endpoint such as /search-tweets or /followers, and send the JSON straight into Google Sheets, Slack, Airtable, or a database. Because the rate limit is a flat 20 requests per second on every plan, a polling Zap will not trip the per-endpoint windows the official API enforces. The catch worth stating up front: it is read-only, so anything that posts, likes, or follows still belongs to the native integration.

Contents

• Does Zapier still work with Twitter/X in 2026?
• What the native X integration can and cannot do
• How to pull X data into Zapier without the native integration
• Zapier Twitter alternatives, and when you actually need them
• Cost: the native path versus a flat-rate read API
• Troubleshooting: Zapier Twitter not working
• FAQ
• Getting started

Does Zapier still work with Twitter/X in 2026? {#status}

Zapier supports the Twitter/X integration as of mid 2026, but on a bring-your-own-key basis: you must connect your own X Developer App, and the X API is billed per use on top of your Zapier plan.

The integration has had a stop-start history. It was removed on August 31, 2023 after Twitter changed its API pricing. Zapier relaunched it in February 2026 with auto-tracking of mentions, keyword monitoring, on-demand search, and publish, like, repost, and follow actions, according to Zapier's own status notes on the X integration. It was then switched off again on March 4, 2026 and restored within days. As of now it is live and functional.

The structural change matters more than the dates. Zapier no longer maintains a shared connection for X. To connect the app you create your own X Developer App and paste its Client ID and Client Secret. That single requirement is what trips most people: the no-code platform now sits on top of a developer account you have to set up and fund yourself.

It also means the native integration costs money beyond Zapier. Read operations on the official API are billed per resource fetched, and polling triggers spend those credits continuously. A trigger like "New Post from Query" that checks every five minutes runs 288 times a day, every day, and each run that returns posts bills per post whether or not the content is new to you. An always-on monitor can quietly accumulate a meaningful X API bill before you have read a single post you care about.

What the native X integration can and cannot do {#native}

The native X app on Zapier is built around live triggers and write actions, not historical data pulls. It fires on new events going forward and cannot reach back in time.

Here is the current surface. Triggers cover New Mention, New Post from Query, and New Post from User. Searches cover Find Post, Find User, and Search Posts. Actions cover Create Post, Follow User, Like Post, Repost, and a flexible API Request step for raw calls.

The gap appears the moment you want data rather than automation of your own posting. A trigger only sees posts published after the Zap is switched on, so it cannot backfill a brand's last six months of mentions or a competitor's tweet history. There is no batch retrieval and no time-range query in the trigger model; you get one new item per fire. This is the same limitation users hit years ago when trying to fetch older tweets or poll metrics over a period, and the relaunched integration did not change it.

For a clean split: the native integration is the right tool when you want to publish to X, react to a brand new mention in real time, or run a small posting workflow. It is the wrong tool when you want to read at volume, pull history, or build a dataset. That second job is what a Twitter mentions API or a search endpoint does directly, and it is where a read-only data source earns its place in a Zap.

How to pull X data into Zapier without the native integration {#pull}

To pull X data into Zapier without the native app, add a Webhooks by Zapier or API by Zapier step, point it at a read API endpoint with your key in the request headers, and map the returned JSON into your destination app. Both steps are premium and need a paid Zapier plan.

Zapier gives you two ways to call an API that does not have a dedicated app, and the choice is about how your key is stored. Per Zapier's guide to making API requests, Webhooks by Zapier holds credentials in plain step fields visible to anyone who can view the Zap, while API by Zapier stores them in a separate connection and injects them automatically. For a static key, API by Zapier with the "Static Headers (API key)" auth type is the more secure option; Webhooks by Zapier works too and is fine for quick builds.

Step by step

1. Get a read API key. With Sorsa API the key is issued instantly and goes in an ApiKey header. No developer app, no OAuth.
2. Choose a trigger. Use Schedule by Zapier to poll on a cadence (every 15 minutes, hourly, daily), or another app's event to kick off the call.
3. Add the API step. Select API by Zapier (auth type Static Headers, header ApiKey: YOUR_KEY) or Webhooks by Zapier with the key added under Headers.
4. Configure the request. Set the method and URL for the endpoint you want, plus the JSON body for POST endpoints.
5. Route the output. Map fields from the JSON response into Google Sheets, Slack, Airtable, Notion, or a database step.

A GET endpoint such as a profile lookup is the simplest case. The raw call looks like this:

curl -H "ApiKey: YOUR_API_KEY" \
  "https://api.sorsa.io/v3/info?username=nasa"

In Zapier that is a Webhooks GET step with the URL above and one header, ApiKey: YOUR_API_KEY.

A search is a POST. The endpoint accepts a query and an order, and returns about 20 tweets per page with a cursor for the next page:

curl -X POST "https://api.sorsa.io/v3/search-tweets" \
  -H "ApiKey: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "your brand", "order": "latest"}'

In Zapier that is an API by Zapier or Webhooks Custom Request step: method POST, URL https://api.sorsa.io/v3/search-tweets, a JSON body with your query and order, and the ApiKey header. The same pattern covers a tweet search workflow, follower pulls through /followers (up to 200 profiles per request), batch profile lookups through /info-batch (up to 100 at once), and trends by location through /trends. The header name and every field map to the live reference, so the call behaves the same in Zapier as it does in a terminal. For the header itself, Sorsa's authentication docs show the exact format.

If you want a trigger that feels native, Webhooks by Zapier has a "Retrieve Poll" mode that makes a GET request to an external URL on a schedule and surfaces new entries, which turns a read endpoint into a near-real-time source. A common build sends every new matching tweet straight to a sheet; if that is your goal, the dedicated walkthrough on exporting X data to Google Sheets covers the field mapping. None of this needs an X Developer App, which is the whole point of the route, and it is the same approach covered in our no-developer-account guide.

Zapier Twitter alternatives, and when you actually need them {#alternatives}

The right alternative depends on whether you are reading or writing. For reading X data, swap the native trigger for a read API call. For posting, you need either the native integration or a connected publishing tool.

If your broken or expensive workflow was about publishing, a scheduler like Buffer or Hootsuite connected to Zapier posts to X on your behalf and handles the X connection for you. Those are the honest answer for write automation, since a read-only API cannot post.

If the workflow was about pulling data, the question is which automation platform you run it on, not whether it is possible. The same read-API-over-webhooks pattern works across the major tools, which is why this is one of a set: the Make version and the n8n version cover the same build on those platforms. Pick the one you already use; the data source does not change.

A quick decision rule. Need to post, like, follow, or react to brand-new mentions in real time? Stay on the native integration (or a publisher). Need to read tweets, profiles, followers, or mentions, pull history, or feed a dataset? Use a read API and skip the developer app entirely.

Cost: the native path versus a flat-rate read API {#cost}

The native bring-your-own-key path costs your Zapier plan plus the official X API's per-resource fees; a flat-rate read API costs your Zapier plan plus a fixed per-request charge. On read-heavy workloads the second is far cheaper.

Work the read math. The official API bills per resource fetched, so 1,000 post reads cost $5.00 at $0.005 each, and pulling the author profiles alongside them adds another $0.010 per profile. The same 1,000 posts pulled through Sorsa's search endpoint on the Pro plan cost roughly $0.10, because about 20 tweets come back per request and each request is billed once, with author profiles included free. Now layer in polling: an always-on monitor checking every five minutes fires 288 times a day, and on the native path each fire that returns posts is billable, while on a flat plan it is one request against a monthly quota.

If a Zap needs to publish, like, or follow, the native integration is the right tool, and that is the official API's territory. For read-heavy pulls at a flat, predictable price with no developer app to maintain, a read-only API such as Sorsa is the cheaper and simpler source. Plans run $49 a month for 10,000 requests, $199 for 100,000, and $899 for 500,000; the full breakdown sits in the 2026 X API pricing comparison.

Troubleshooting: Zapier Twitter not working {#troubleshooting}

Most "Zapier Twitter not working" reports trace to the integration being switched off or a connection problem, not a bug in your Zap.

The signature failure is a 401 error reading "Could not authenticate you" with code 32. After the 2023 removal and the March 2026 outage, that error simply meant the integration was unavailable. Since the relaunch, the most common cause is a missing or misconfigured X Developer App, since Zapier now requires your own Client ID and Client Secret rather than a shared connection.

Other recurring issues are narrower. Tweets that report as sent but never appear on X usually point to an expired connection or a content policy block on the post text. Posts that exceed the character limit fail or truncate, so a Formatter step to trim text helps. And because the native triggers poll the X API, heavy schedules can run into the official rate windows, which a read API on a flat 20 requests per second avoids. If the workflow that broke was read-only, re-pointing it at a webhook call to a data API sidesteps the auth chain entirely, since there is no X login in the loop.

A team that swapped its broken Twitter zaps for a read API

A small marketing analytics team, around eight people, had run brand-mention monitoring through Zapier's native Twitter trigger for two years: every matching mention landed in a shared sheet and pinged a Slack channel. When the integration went dark in early 2026 and came back demanding a self-funded X Developer App, the per-resource billing on continuous polling made the rebuild unattractive.

They kept the Zap shell and replaced only the data step, swapping the native trigger for a scheduled Webhooks call to a search endpoint with the key in an ApiKey header. The Slack and Sheets steps downstream did not change. The flat per-request billing cut their data cost by roughly 30 to 50 times against the official API's per-resource model, the predictable monthly number that anyone moving a read-heavy workload off the official API tends to see. No developer app, no OAuth, no surprise invoice from polling.

FAQ {#faq}

Does the Twitter/X integration still work on Zapier in 2026?

The Twitter/X integration works on Zapier as of mid 2026, but on a bring-your-own-key basis. After being removed in August 2023 and briefly switched off again in early March 2026, it was restored with a requirement that you connect your own X Developer App using its Client ID and Client Secret. Zapier no longer provides a shared connection, and the X API is billed per use on top of your Zapier plan.

Can you pull historical tweets into Zapier?

The native X triggers on Zapier only fire on new posts going forward, so they cannot backfill tweets posted before the Zap was switched on. To pull historical or older tweets into a Zap, call a search or user-timeline endpoint from a read API through a Webhooks by Zapier step, which returns past posts on demand rather than waiting for new ones to arrive.

Do you need an X Developer App to get X data into Zapier?

You need an X Developer App only for the native X integration. To read X data without one, point a Webhooks by Zapier or API by Zapier step at a third-party API. Sorsa, for example, authenticates with a single key in an ApiKey header and returns tweets, profiles, followers, and mentions, so there is no developer app to create and no OAuth flow to maintain.

How much does it cost to read X data through Zapier?

Reading X data through Zapier costs your Zapier plan plus whatever data source you use. The native integration adds the official X API's per-resource fees, where 1,000 post reads run about $5.00. A flat-rate API such as Sorsa instead bills one request per call regardless of how many tweets return, with plans from $49 a month for 10,000 requests, which works out far cheaper on read-heavy workloads.

Why did your Zapier Twitter integration stop working?

Most Zapier Twitter failures since 2023 trace back to the integration being removed or switched off, which surfaces as a 401 error reading "Could not authenticate you." After the 2026 relaunch, the most common cause is a missing or misconfigured X Developer App, since Zapier now requires your own Client ID and Client Secret. Tweets that report as sent but never appear usually point to an expired connection or a content policy block.

Can Zapier post tweets automatically?

Zapier can post tweets automatically through the native X integration's Create Post action, which requires your own X Developer App, or through a connected scheduler like Buffer that posts to X on your behalf. Read-only data APIs cannot publish posts, like, follow, or repost, so any workflow that writes to X needs the native integration or a social media management tool.

Getting started {#start}

If your goal is reading X data inside Zapier, the fastest start is a key and a single webhook step. Sorsa issues a key in about three minutes, runs on a flat 20 requests per second across every plan, and bills one request per call with author profiles included free. Try endpoints in the API playground before you wire anything up, then check plan limits on the pricing page. For read-heavy monitoring and dataset workflows, the flat model is the cheaper, more predictable source; for posting, the native integration remains the path.

Reviewed by Keksich, founder of Sorsa, marketer and X API researcher.

This guide draws on our day-to-day work operating Sorsa's read API and on Zapier's published status for the X integration, verified on June 17, 2026. Sources include Zapier's status notes confirming the bring-your-own-key relaunch and its documentation on Webhooks by Zapier versus API by Zapier. Official X API figures reflect the pay-per-use rates in effect after the April 20, 2026 change. The cost comparison weighs the native developer-app path against a single flat-rate read source.