Notifications

Review alerts from your AI secretary, new leads, and booking requests.

Open Notifications to review alerts such as new leads, booking requests, escalations from the AI Receptionist, and other account activity. Treat the list as a short queue for decisions, not a place to let messages age silently.

What each alert means

• New lead — follow up, qualify the request, and decide the next step.
• Booking request — approve, decline, or request more detail.
• AI escalation — call the customer back or assign the handoff.
• Payment or invoice event — confirm the payment, send a reminder, or review the account.
• Channel activity — check messages that need a human reply.

Turn alerts into actions

Mark items as read as you handle them so the list stays actionable. Use the linked record to update the CRM note, change a job status, or send a reply instead of closing the notification without follow-through.

Prioritize AI handoffs

When the AI hands off an angry or sensitive call, the notification is your cue to call the customer back quickly. Open the transcript summary first so you can acknowledge the issue and pick up the conversation with context.

Adjust delivery preferences

Check preferences in account settings if you want email or in-app delivery adjusted. Choose the channels that match how your team actually works, and keep at least one person responsible for urgent alerts every day.

Tip: Notification tooling depends on your plan; Pro and above include the richer ops notifications add-on.

Build a morning alert review

Start the day by sorting notifications into three groups: requests to approve, leads that need a human reply, and handoffs where a customer is waiting. Handle the handoffs first, then approve or decline booking requests, then work the lead queue. Clearing the list in this order prevents the most urgent customer experience from waiting behind routine activity.

Close the loop on every alert

When an alert links to a customer, job, or booking request, open that record and complete the next action before marking the alert done. A notification is not resolved just because it was read; it is resolved when the customer, invoice, or schedule has been updated.

Review delivery quality weekly

Once a week, ask whether any urgent alert was missed and whether the team is receiving too many or too few messages. Change delivery preferences for the people who are responsible for each type of alert, and test the new setup with a real or test event before relying on it.

What to do when an alert is missing

If an expected alert does not arrive, check the linked record first instead of assuming the customer was never contacted. Open the customer, job, or invoice to see whether the event happened and whether the alert was delivered to the right user. Then review delivery preferences and confirm the responsible person is still active on the account. Keep at least one backup owner for urgent categories so a vacation day does not leave the queue unattended.

Build a simple priority matrix

Assign a priority to each alert type so the team does not sort the queue from memory. A customer waiting on an AI handoff is urgent and is handled immediately. A booking request is important and should be approved or declined before new leads are worked. A routine lead, channel message, or payment event can be handled in batches. Write the matrix in the team's normal words and post it near the dashboard workflow.

Name an owner for each alert

Every alert category needs one primary owner and one backup. The owner decides whether the alert needs a reply, a schedule change, or a CRM update. The backup covers vacations and busy days. If an alert is constantly unowned, change the delivery preference or add a second owner before a customer is left waiting.

Turn notifications into a countable workflow

At the end of each day, note how many alerts arrived, how many were resolved, and which ones carried over. A growing carryover list means the workflow is creating work faster than the team can close it. Reduce the noise or move more of the queue to a specific owner instead of accepting an ever-increasing backlog.

Write response-time buckets

Set three simple response expectations: a customer waiting on an AI handoff is handled the same day, a booking request is approved or declined on the requested date or before the next business day, and routine leads are worked in a scheduled batch. Write the buckets near the dashboard and use them when a notification waits too long. The team does not need to invent a priority every time an alert appears.

Decide whether an alert needs a call

For each alert, decide whether it can be closed with a message or requires a phone call. A customer waiting after an AI escalation usually needs a voice call. A booking request can often be handled with a message when the request is clear. A routine lead can be handled in the preferred channel of the customer. This decision keeps the queue from growing while messages wait for a convenient time.

Keep a one-line daily handoff

At the end of each day, write one line for any alert that carries over: what it is, who owns it, and when it must be resolved. Put that line where the next shift starts. A short handoff prevents the morning review from starting with stale context or an unread notification from the previous day.

Keep one weekly digest for learning

Once a week, read the alerts that were resolved and list the categories that needed a callback. If the same type of alert repeatedly needs a phone call, update the workflow so the responsible person gets it sooner. This short digest shows whether the notification setup is actually reducing work or simply moving it to a different queue.

Create a notification response worksheet

For each alert type, write the owner, the response time, the action to complete, and the record to update. A booking request owner approves or declines and updates the job. An AI handoff owner calls the customer and records the outcome. A lead owner qualifies the lead and sets the next follow-up. A payment event owner confirms the invoice and updates the status. The worksheet turns the notification list from a message board into a decision queue.

Common notification problems and fixes

• Too many alerts: reduce delivery preferences or assign one owner to batch routine items.
• Too few alerts: confirm the responsible user is active and notification preferences still match the role.
• Alerts missed on a day off: name a backup owner for urgent categories.
• A notification is read but never resolved: link it to a record and complete the action before clearing it.
• The same alert appears twice: check whether a duplicate record or duplicated workflow created the event.

Test notification delivery monthly

Once a month, trigger a test booking request, AI handoff, and invoice event if the environment allows it, then confirm each alert reaches the intended owner. This check catches expired email addresses, inactive users, and outdated delivery preferences before a real customer event reveals them.