Pipedrive webhooks
Last verified against Pipedrive's documentation on September 27, 2026.
Pipedrive sends webhooks when something changes in a Pipedrive company: a deal is created, a person is updated, an activity is deleted. Pipedrive’s delivery rules are strict. Failed first attempts count toward a ban that silently drops events, and a webhook with no successful deliveries for three days is deleted outright. Pipedrive also doesn’t sign webhooks. It uses HTTP Basic Auth instead, which changes how you secure a forwarding setup. This page covers what Pipedrive sends, how it delivers, how to fix the common problems, and where HookDeploy fits.
What Pipedrive sends webhooks for
A Pipedrive webhook combines an action and an object, written action.object: for example create.deal, change.person, or *.* for everything.
| Part | Values |
|---|---|
| Actions | create, change, delete, or * for all |
| CRM objects | deal, lead, person, organization, activity, note, product, deal_product, deal_installment |
| Pipeline setup | pipeline, stage |
| Projects | project, board, phase, task |
| Users | user |
Objects refer only to themselves. To hear about a deal’s owner changing, subscribe to change.deal, not user.
Every webhooks v2 payload has three parts:
| Block | On create | On change | On delete |
|---|---|---|---|
meta | Event details | Event details | Event details |
data | The new object | The object’s current state | null |
previous | null | The previous values of the fields that changed | The object’s last state |
The meta block includes the event id, action, entity, entity_id, attempt, timestamp, change_source (app or api), is_bulk_edit, and the webhook_id.
Version 2 is the default for new webhooks. Version 1 payloads have a different shape, so check the version field in meta if you’re migrating.
How Pipedrive delivers webhooks
| Behavior | Detail |
|---|---|
| Method and format | POST with a JSON body |
| Authentication | Optional HTTP Basic Auth username and password, sent in the Authorization header. No signature |
| Success | Any 2xx |
| Timeout | 10 seconds |
| Retries | 3 more attempts, after 3, 30, and 150 seconds. The meta.attempt field shows which attempt it is |
| Ban | Each failed first attempt adds 1 to a ban counter. At 10, the webhook is banned for 30 minutes and events during the ban are lost |
| Deletion | A webhook with no successful deliveries on 3 consecutive days is deleted |
| Redirects | The subscription URL must not redirect |
| Permissions | Events are checked against the webhook’s user. Use a top-level admin’s user to receive everything |
| Limits | 40 webhooks per user |
| Rate limits | Outgoing webhooks don’t count against Pipedrive’s API rate limit |
Common Pipedrive webhook errors and how to fix them
Events missing for 30 minutes at a time
That’s the ban system. Every time a first delivery attempt fails, by returning a non-2xx status or taking longer than 10 seconds, the webhook’s ban counter goes up. At 10, Pipedrive bans the webhook for 30 minutes, and events during the ban are dropped, with no retries. Fix slow or failing responses first: respond with 2xx immediately and do the work from a queue.
Webhook deleted
If a webhook has no successful deliveries on 3 consecutive days, Pipedrive deletes it. After an outage or a broken deploy, check Settings, then Tools and apps, then Webhooks in Pipedrive, and recreate anything that’s gone. Then backfill the gap through the API.
Missing events for some records
Events are checked against the permissions of the webhook’s user. If that user can’t see a deal or person, no event is sent for it. Create the webhook with a top-level admin’s user_id to receive events for everything.
Basic Auth rejected
Pipedrive sends the username and password you set on the webhook as a standard Authorization: Basic header. Compare it in constant time. If there’s a proxy in front of your handler, make sure it passes the Authorization header through.
Duplicate events
Retries resend the same event, and your handler may have finished its work but responded too late. Deduplicate on meta.id. The meta.attempt field tells you it’s a retry, but don’t rely on it alone.
Infinite loops with your own integration
If your handler writes back to Pipedrive, those writes trigger new webhooks. Check meta.change_source: skip events marked api if they came from your own integration’s writes.
Subscription URL rejected
Pipedrive rejects subscription URLs that redirect. Register the final URL exactly as your server serves it, including the scheme and any trailing slash.
Why use HookDeploy with Pipedrive
No more bans or deleted webhooks
Pipedrive’s ban counter only counts failed first attempts, so a handler that’s slow for a few minutes can get a webhook banned and lose 30 minutes of events. HookDeploy captures each delivery at Cloudflare’s edge and responds to Pipedrive right away, before forwarding it to your server. First attempts succeed, the counter doesn’t climb, and an extended outage on your side can’t get the webhook deleted. From there:
- Destination retries. Starter retries 3 times (after 30 seconds, 5 minutes, and 30 minutes). Pro and Scale retry up to 5 times on a schedule you choose. Free doesn’t retry automatically.
- Incidents. Forward failures are grouped into incidents, and you can bulk-replay them once your server recovers.
- History. Pipedrive’s retries end within about three minutes. HookDeploy keeps captured deliveries for your plan’s retention period, from 7 days on Free to 365 days on Enterprise.
Two authentication hops, each done properly
With HookDeploy in the middle, there are two connections to secure:
- Pipedrive to HookDeploy. Add HookDeploy’s endpoint token to the URL you give Pipedrive as
?token=. HookDeploy rejects requests without it before they’re captured. Endpoint tokens are available on Starter and above. - HookDeploy to your handler. Inject an
Authorizationheader on the forward destination, with a secret only your handler knows. HookDeploy adds it to every forward, including retries and replays to that destination.
That second hop matters, because of how HookDeploy stores requests. See Authentication on retries and replays.
One webhook, many systems
Pipedrive allows 40 webhooks per user. Point one at HookDeploy and forward each event to your app, a data warehouse, and a team channel in parallel. A failing destination doesn’t hold up the others. Free includes 1 destination per endpoint, Starter includes 5, and Pro and above are unlimited.
Reshape payloads for other tools
Pipedrive payloads carry the whole object. For destinations that only need a few fields, such as a Slack alert on a won deal, HookDeploy’s transformations can pick out the fields you need, rename them, and format values. Transformations are available on Starter and above.
Sync to internal systems
The HookDeploy Agent connects outward from your network over mTLS and delivers events to a local port, so an internal database or reporting service can follow Pipedrive without a public endpoint.
Know the limits
- Plan limits. On Free, requests beyond the monthly allowance or the daily limit get a
429. Pipedrive counts that as a failed delivery, and a failed first attempt counts toward the ban. Paid plans keep accepting traffic past the included volume. - Tokens in the URL. Pipedrive can’t send custom headers, so the endpoint token goes in the webhook URL, where Pipedrive users who can see webhook settings can see it. HookDeploy removes the token before forwarding, so your handler never receives it.
Authentication on retries and replays
HookDeploy redacts the Authorization header from stored requests by default, to keep credentials out of your request history. The first forward uses the original request, so it carries Pipedrive’s Basic Auth header. But retries and replays are sent from the stored copy, where that header has been redacted.
If your handler checks Pipedrive’s Basic Auth credentials, the first delivery passes and every retry and replay fails. So don’t authenticate HookDeploy’s forwards with Pipedrive’s credentials. Inject your own Authorization header on the destination and check that instead. An injected header replaces a matching header from the original request, and HookDeploy adds it to every forward, every retry, incident replays, and manual replays you send to that destination. A manual replay to a custom URL doesn’t use destination settings, so add the header as a one-time header on that replay.
Duplicates are the other thing to handle. Pipedrive’s retries, HookDeploy’s retries, and replays can all deliver the same event. Deduplicate on meta.id.
Set up Pipedrive with HookDeploy
- In HookDeploy, create an endpoint (for example,
Pipedrive production) and copy its URL. - On the endpoint’s Security tab, under Token, generate an endpoint token (Starter and above).
- On the Forwarding tab, add a destination pointing to your handler, or to an agent for private delivery. Add a header
Authorization: Bearer YOUR_HANDLER_SECRETand set the destination’s retries. - In Pipedrive, go to Settings, then Tools and apps, then Webhooks, and create a webhook. Use your HookDeploy URL with
?token=YOUR_ENDPOINT_TOKENadded, choose the event action and object, and pick the user whose permissions it should use. - Create or edit a record covered by the webhook, and confirm the event appears in HookDeploy and reaches your handler.
Example handler (Node.js and Express)
import crypto from 'node:crypto';
import express from 'express';
const app = express();
const expectedAuth = Buffer.from(`Bearer ${process.env.HANDLER_SECRET}`);
app.post('/webhooks/pipedrive', express.json(), async (req, res) => {
// Check the header HookDeploy injects on every forward, including retries and replays to this destination.
const received = Buffer.from(req.get('authorization') ?? '');
const valid =
received.length === expectedAuth.length &&
crypto.timingSafeEqual(received, expectedAuth);
if (!valid) return res.status(401).send('Unauthorized');
res.sendStatus(200);
const { meta, data, previous } = req.body;
if (meta.change_source === 'api' && isOwnWrite(meta)) return; // avoid loops
await enqueue({ eventId: meta.id, action: meta.action, entity: meta.entity, data, previous });
// Deduplicate on meta.id in your worker.
});
FAQ
Does Pipedrive sign its webhooks?
No. Pipedrive webhooks carry no signature. You can set an HTTP Basic Auth username and password on a webhook, and Pipedrive sends them in the Authorization header of every delivery for your endpoint to check.
Why did my Pipedrive webhook stop sending events for 30 minutes?
Pipedrive bans a webhook for 30 minutes once 10 first attempts have failed, and events that occur during the ban are lost for good. HookDeploy responds to Pipedrive right away even when your server is slow or down, so first attempts don't fail and the counter doesn't climb.
Why was my Pipedrive webhook deleted?
Pipedrive deletes a webhook that has no successful deliveries on 3 consecutive days. With HookDeploy in front of your server, deliveries succeed as long as HookDeploy captures them, so an extended outage on your side doesn't cost you the webhook.
Why do HookDeploy retries fail my Basic Auth check?
HookDeploy redacts the Authorization header from stored requests, and retries and replays are sent from the stored copy, so Pipedrive's Basic Auth credentials aren't on them. Authenticate HookDeploy's deliveries with a header you inject on the destination instead. HookDeploy adds it to every forward and retry, and to replays sent to that destination.
How do I tell webhook retries apart from new events?
Every Pipedrive payload has a meta.id identifying the event and a meta.attempt counter showing which delivery attempt it is. Deduplicate on meta.id, since the same event can arrive more than once.
How do I avoid reacting to changes my own integration made?
Check meta.change_source. It's app when a change came from the Pipedrive web app and api when it came through the API, which lets an integration skip events caused by its own writes.