SupportConfigure Ask360 auto recommendations
Schedule evidence-based bot recommendations with a workday window, interval, cooldown, and approved read-only signal playbooks.
How this fits into Lead360Explore Ask360See how Pieces, memory, channels, playbooks, learning, and kits work together.- 01
How Lead360 helps
- 02
Before you begin
- 03
Complete the workflow
- 04
Understand the key controls
How Lead360 helps
Auto recommendations lets an Ask360 bot review approved read-only signals and surface a useful next action when evidence supports it. Workday windows, intervals, cooldowns, and signal playbooks limit noise and prevent the feature from manufacturing filler when there is no meaningful recommendation.
When to use this workflow
Use Auto Recommendations when a bot should periodically identify review-worthy work from approved signals without carrying out the proposed action automatically.
Before you begin
Start with the exact sub-account and the smallest set of permissions, records, and resources required for the task.
- Choose a bot with a clear operational owner.
- Create and test read-only signal playbooks.
- Define workday timezone, active window, interval, cooldown, and what counts as useful evidence.
Complete the workflow
Follow the steps in order and stop when a required resource, permission, review, or approval is missing.
- Open the bot profile and Auto Recommendations settings.
- Set the workday timezone and active hours.
- Choose the review interval and cooldown.
- Assign approved read-only signal playbooks.
- Run controlled observations, then enable recommendations and review their evidence and usefulness.
Understand the key controls
Read-only signals
Signal playbooks should observe and summarize. A recommendation does not approve the action it proposes.
No filler
The bot can return no recommendation when evidence is insufficient. Frequency should not force content.
Cooldown
Cooldown limits repeated recommendations from the same conditions and should be tuned with the business workflow.
Confirm the result
Do not rely on a success message alone. Confirm the saved state and any live effect in the product area that owns the final record.
- Confirm recommendations appear only during the intended window.
- Open the cited records and validate the evidence.
- Track dismissals, duplicates, stale items, and follow-up outcomes before increasing frequency.
Troubleshooting
Recommendations are repetitive
Increase cooldown, narrow signal playbooks, deduplicate stable conditions, and require stronger evidence.
No recommendations appear
Check schedule timezone, enabled state, signal playbook results, bot resources, and whether no useful evidence is the correct result.
Frequently asked questions
Do auto recommendations perform the suggested action?
No. They surface an evidence-based suggestion. Any live operation remains subject to normal user, capability, effect, and confirmation controls.
Should the bot always produce a recommendation?
No. Returning no recommendation is correct when approved signals do not support a useful next action.
Continue with Lead360
Use this workflow in Lead360
Put this guide into practice with a 14-day free trial of Lead360 Pro early access, or continue learning through the related collection.