WhatsApp us
Worked example (illustration)

Worked example: a support triage agent with a stop rule

A filled-in stop rule for a fictional D2C brand's support agent, showing exactly how we decide what an agent does alone and when it hands over.

Talk on WhatsAppTake the pilot scorecard
On this page (13 sections)
  1. Who is the fictional business?
  2. What is the agent's purpose?
  3. Which systems can it use?
  4. What may the agent do alone?
  5. What needs approval?
  6. What are the hard stops?
  7. What does the handover include?
  8. How is the stop rule reviewed?
  9. What does this example show?
  10. How do you know whether the pilot worked?
  11. What should you prepare before talking to us?
  12. Related reading
  13. Questions

Who is the fictional business?

Illustration only: "Kaapi & Co." is a fictional Bengaluru D2C coffee brand created for this example. It is not a client, and no results are claimed.

Kaapi & Co. sells coffee and brewing equipment online. Support questions arrive on WhatsApp and email: order status, subscription changes, grind recommendations, damaged deliveries and the occasional angry message. A team of two handles everything.

What is the agent's purpose?

To triage every incoming message, answer routine questions from live data and approved content, draft replies for the rest, and make sure nothing urgent waits in the queue.

Which systems can it use?

SystemReadWriteLimits
ShopifyYesNoOrder status, items, tracking only
Subscription appYesNoNext delivery date and plan
HelpdeskYesYesTags, priority, internal notes, drafts
Knowledge baseYesNoApproved help articles and brewing guides

Never ask for or repeat: full card numbers, UPI PINs or OTPs, passwords.

What may the agent do alone?

  • Answer order status and tracking questions from Shopify
  • Answer brewing, grind and product questions from approved guides
  • Share the next subscription delivery date
  • Tag and prioritise every ticket
  • Send a holding reply ("We've received this; the team will respond by [time]") for tickets in the approval queue

What needs approval?

ActionApproverTarget time
Replies to damaged or wrong-item reports (replacement or refund)Support leadSame working day
Subscription pause, skip or cancel confirmationsSupport agentSame working day
Replies to wholesale or café enquiriesFounderWithin 2 working days

What are the hard stops?

  • Customer asks for a person
  • Anger, abuse, or a third message without resolution
  • Words like "fraud", "chargeback", "consumer court", "lawyer"
  • Health concerns (for example, a reaction after drinking)
  • Any request about personal data (access, deletion)
  • Agent unsure of intent after one clarifying question

Stop message: "Thanks for telling us. I've passed this to our team, and someone will reply personally by [time today/tomorrow]."

What does the handover include?

The full conversation, Shopify order link, tags, the reason for stopping and the agent's suggested next step (labelled as a suggestion), posted to the helpdesk with high priority and a WhatsApp alert to the support lead during working hours.

How is the stop rule reviewed?

Weekly for the first month: the support lead reviews all hard-stop handovers and a sample of autonomous replies. Any change, such as moving subscription skips to "alone", is written in the change log and approved by the founder.

What does this example show?

Most messages can be handled or drafted automatically, while every decision involving money, health, legal risk or emotion stays with a person. You can write the same document for your own workflow with our free template.

Our stop rule

The agent drafts, routes and escalates. A person decides. Before anything is built we write down what the agent may do alone, what needs approval, and when it must stop and hand over with the full context attached.

How do you know whether the pilot worked?

We agree the success measure before building, and it should be something your team can count without a dashboard project. Typical measures include the share of enquiries logged the same day, time to first response, the number of follow-ups that actually happened, hours spent on the task per week, or the number of escalations that were correctly flagged. We count the baseline before launch (even by hand for a week) and again after a few weeks of running.

We also review the qualitative side with the people involved. Do they trust the drafts? Are handovers arriving with enough context? Are customers asking for a person more or less often? A pilot can hit its number and still need tuning, or miss it and teach you something important. Either way, the decision to keep, tune, expand or stop is yours, made on evidence. We don't publish generic success figures, because results depend on your process, volumes and team.

What should you prepare before talking to us?

You don't need a technical brief. A useful first conversation needs only a few things, and gathering them is valuable even if you never work with us:

  • The process in plain words: what triggers it, who does what, and where it usually goes wrong.
  • Rough volumes: how many messages, orders, tickets or invoices per week. A one-week tally is enough.
  • Your tools: WhatsApp app or API, CRM, sheets, accounting or helpdesk software, including versions or plans where you know them.
  • Three real examples: typical messages or cases, plus one awkward one, with personal details removed.
  • What must never happen: the mistakes that would hurt most. These become the first lines of your stop rule.
  • An owner: the person who will review the workflow once it runs.

Send these on WhatsApp or by email, or run the pilot scorecard first.

Questions

Is Kaapi & Co. a real company?

No. It is fictional and used only to illustrate a stop rule.

Would you start with this full scope?

Often we'd start smaller, for example triage and order status only, and add drafting after review.

Can this work for SaaS support?

Yes, with different hard stops. See SaaS and B2B.

Next step

Start with one workflow

Tell us which process eats your team's time. You'll get an honest view of whether it's worth automating and what a first pilot would look like.

WhatsApp +91 99442 26010Email us
WhatsApp us
WhatsAppCall