Telltale, by GTM Cloud · Watch · Early access
Salesforce error monitoring that knows when something breaks, and when it goes quiet
Telltale reads the logs your org already writes, groups repeated errors into issues, and runs the action you chose for the ones you recognize. It also notices when a job that should report in goes silent.
You know this one
The message lands at 8:05. A sales manager, polite and a little puzzled: the new accounts from the billing system aren’t showing up. They haven’t since Monday.
You check. The nightly job failed Monday, and again Tuesday, and again Wednesday. Same error, written to a log object each time, row after row. Nobody was looking at that object. Why would they?
That’s the tidy version, where the job at least wrote an error. The worse one is the job that stopped running and wrote nothing at all.
Salesforce error monitoring that tells the right team first
Telltale turns the logs your org already writes into one common shape. Repeated errors are grouped into issues with a count, so a thousand rows become one line you can work. Rules recognize the errors you know and run the action you chose: a Flow, Apex, a notification, an email or a platform event.
For the jobs that must not stop, you add a heartbeat. If the signal doesn’t arrive in time, Telltale raises an issue.
A telltale is the strip of yarn on a sail that shows a problem with the wind before you feel it.
See it first. Take it back. Keep the receipt.
| Preview | Undo | Prove |
|---|---|---|
| Try it shows what a rule matches, and a new action starts as a dry run. | Reopen an issue at any time; actions are held back by a throttle, a daily cap and a circuit breaker. | Every event, issue and action run is kept, for the retention you set. |
What you get
It reads what you already log
Each source maps your log rows, emails or events onto one common shape, with a severity map.
One issue, not a thousand rows
Lines with the same fingerprint become one issue with a count. Work the list from Home: acknowledge, resolve, ignore or reopen, one at a time or in bulk.
Rules that act
A rule is a pattern plus conditions. Try it shows what the rule matches before you rely on it, and you attach a Flow, Apex, a notification, an email or a platform event.
It notices silence
A heartbeat is a signal you expect on a schedule. When it doesn’t arrive, Telltale raises an issue, so a stopped job isn’t found a week later.
Guarded, so it doesn’t flood people
Every action can start as a dry run and has a threshold, a throttle, a daily cap and a circuit breaker. Redaction runs as lines come in.
Yours to switch on
The install starts nothing. A source needs a clean test run before it’s enabled, and the engine runs only after you choose Start engine.
How it works
- Assign access. Builders get Telltale Admin, issue handlers Telltale Analyst, an integration user Telltale API.
- Add a source. Use a template or add your own on the Setup tab, choose Test run, then enable it.
- Start the engine. Read the Health Check, set the admin email and choose Start engine.
- Write rules. Copy a starter or write one, run actions as dry runs first, and add heartbeats.
Two readers, two reasons
If you’re the admin
Point Telltale at a source with a template, test-run it and switch it on. Then write rules, attach the action, and add heartbeats for the jobs that must not stop. Nothing runs until you start the engine.
If you’re the one signing off
Most orgs find out about a failed integration from the person it failed for. Telltale tells the right team first, about the errors you know, the ones you don’t, and the nightly job that never ran.
Today, and with Telltale
| Today | With Telltale |
|---|---|
| Errors spread over objects, emails and events | One list of issues, with counts |
| The same error a thousand times in a list | Known errors handled by a rule and an action |
| Users report failures before the team knows | The right team notified first |
| A stopped job found days later | A heartbeat for every job that must not stop |
Where it’s strong, and what to know first
We’d rather you find the limits here than after you install.
Where Telltale is strong
- Reads the logs you already have: a logging object, Flow and Apex job failures, error emails, the API and platform events
- Repeated errors become one issue with a count, not a thousand rows
- Rules run an action: a Flow, Apex, a notification, an email or a platform event
- Heartbeats raise an issue when an expected signal doesn’t arrive
- Guards on every action: dry run, threshold, throttle, daily cap and circuit breaker
Good to know before you install
- Telltale reads logs. It isn’t a logging framework, so your org still needs something that writes them, or the platform’s own error events and emails.
- Errors no rule recognizes are grouped into issues for a person to triage. Telltale doesn’t diagnose them for you.
- Reading error emails needs an Email Services address that an admin creates. Catching another team’s platform event needs a small flow, which the app writes for you to deploy.
- Polled sources are read every few minutes, not in the same instant.
Questions people ask
How do I get alerted when a Salesforce integration fails?
Point Telltale at wherever the integration writes its errors: a logging object, an error email, the API or a platform event. Write a rule for the error and attach an action, such as a notification or an email. Add a heartbeat so a run that never happens raises an issue too.
Can I get notified when a Flow or Apex job fails?
Yes. Source templates read Flow errors and failed Apex jobs, so you don’t have to change how those are logged. Repeats are grouped into one issue with a count, and a rule can send a notification or an email, or run a Flow or Apex.
Do we have to change how we log?
No. Telltale maps the logs you already write onto one common shape, so your existing logging stays as it is. Templates cover a common logging framework, Flow errors and failed Apex jobs, and you can add your own sources for anything else.
Will it flood people with alerts?
It’s built not to. Repeats are grouped into one issue, and every action has a threshold, a throttle, a daily cap and a circuit breaker. A new action can also start as a dry run, so you see what it would do before it does anything.
What about sensitive data in log lines?
Redaction runs as lines come in, and retention is yours to set. You choose how long events, issues and action runs are kept. Everything stays inside your org, so nothing is sent to an outside service to be stored or read.
What is a heartbeat?
A heartbeat is a signal you expect on a schedule, such as a nightly job reporting success. If it doesn’t arrive in time, Telltale raises an issue. Silence is often the first sign of a failure, and a stopped job writes no error.
Works well with
- Portage: Portage loads files a person brings to it. Telltale watches the integrations that run without anyone.
- Wake: when an issue points at a record, field history shows what changed on it.
Telltale is part of the Prove stage of the lead-to-renewal map. Want help setting up sources and rules? Our Salesforce services can help. Every app installs by itself.
Built on the platform
- Native to Salesforce. Sources, rules, actions and heartbeats are Custom Metadata, and events and issues are records in your org.
- Three permission sets: Admin, Analyst and API, the last for an integration user that only pushes log lines.
- In-app guides in English, German, Spanish, French (Canada) and Portuguese (Brazil).
Name the failure you heard about from a user
Ask for early access and a walk-through on your own logs. Connect one source, write one rule, and add a heartbeat for the job you worry about.