WhatsApp helpdesk and ticketing: a guide for Indian teams
A WhatsApp helpdesk turns each customer issue into a ticket with an owner, a priority and a reply deadline, instead of a chat thread somebody has to remember. It matters once you pass roughly thirty conversations a day, because that is the point at which a shared inbox starts losing things quietly.
- A conversation never ends; an issue does. That difference is what makes a ticket useful.
- Shared inboxes fail silently, which is why teams notice too late.
- Reply deadlines must be counted in working hours or they are meaningless.
- Reopen rate is the honest test of whether your team is resolving or just replying.
The point where a shared inbox stops working
Every Indian business starts the same way. One WhatsApp number, two or three people who can see it, and a rule that whoever gets to it first replies. For a while this is genuinely the right answer. It is fast, there is no software to learn, and the team can see everything.
The failure is gradual and quiet. Two people reply to the same customer with different answers. A message that needed the warehouse gets read, mentally filed, and never picked up. Somebody goes on leave and the four conversations they were handling stop moving, because nobody knew they were handling them.
Nothing breaks visibly. There is no error and no alert. You find out a week later when a customer writes "I asked about this on Tuesday" and nobody can say what happened on Tuesday.
The rough threshold is thirty conversations a day, or three people sharing an inbox. Below that, memory works. Above it, memory is the system, and memory does not scale.
There is a version of this that looks fine from the outside for months. Replies still go out, customers still get answers, and the team feels busy rather than broken. What has actually happened is that the inbox now depends on one or two people holding the whole picture in their head, and the business has quietly become fragile in a way no dashboard shows.
What a ticket adds that a chat thread cannot
The distinction people miss is that a conversation and an issue are different objects. A WhatsApp thread with a customer runs for years, across a dozen unrelated problems. An issue starts, gets worked, and finishes.
A ticket is the issue. Because it has a beginning and an end, you can finally measure things that are meaningless against a thread:
- First response time. How long the customer waited for a human, per issue.
- Resolution time. How long the issue took to finish.
- Reopen rate. How often "resolved" did not stay resolved.
- First-time fix. How often an issue was closed without coming back.
None of those can be computed against a conversation that never ends. That is the whole argument for ticketing, and it has nothing to do with software preference.
The second thing a ticket adds is an owner. A named person is responsible, and when they go on leave that responsibility can be handed over rather than evaporating.
It also changes what a handover means. In a shared inbox, handing over work means describing it, and the description is always thinner than the reality. With a ticket, the history is the handover: what the customer said, what was tried, what the warehouse replied, all in one place and in order.
What it looks like on Convo360
The Support Desk is a paid add-on, switched on per workspace. Once it is on, a ticket opens by itself the first time a customer writes in on any channel: WhatsApp, Instagram, website chat, SMS, Telegram or a connected email mailbox. Nobody has to remember to create one.
Each ticket carries a number, a priority, an owner, internal notes the customer never sees, and the full history of what changed and when. If the customer replies within seven days of a resolution, the same ticket reopens rather than a new one appearing, which is what makes the reopen count mean something.
Around that sit the parts a support manager actually runs the team on: routing rules, reply deadlines counted in your own working hours, escalation when a deadline is missed, satisfaction surveys on the channel the customer used, and a dashboard that answers the questions a weekly review asks. There is more detail in the live chat docs.
Because Convo360 keeps the desk and the inbox in the same place, your agents do not learn a second tool. They keep replying in the same conversation view they already use, and the ticket updates itself around them. The parts that need a decision, such as priority or owner, sit beside the conversation rather than in a separate system.
One detail worth knowing early: the desk does not stop the AI working. The bot keeps answering what it can, and the ticket records that it did. Most teams find the majority of tickets close without a human ever opening them, which changes what the worklist is for. It stops being everything that happened and becomes only the things that needed judgement.
Getting the first week right
Most teams over-configure on day one and abandon half of it by day ten. A better order:
- Switch it on and change nothing. Work a week of real tickets with the defaults.
- Set your working hours and timezone honestly, including Saturday if you work it.
- Add reply deadlines you are already meeting, not the ones you wish you met.
- Only then add routing, tags and surveys.
The reason for that order is credibility. A deadline that is missed every day trains the team to ignore the alert, and an alert everybody ignores is worse than no alert, because it is now noise on top of the work.
Keep the tag list short at first. Five or six tags that describe what a customer was upset about will tell you more than thirty that describe your internal process.
Resist the urge to build a status for every situation. Four statuses cover almost every team: open, waiting on the customer, resolved and closed. Teams that add six more spend their review meetings arguing about which one applies rather than about the customers.
Similarly, leave routing off in week one. Watch who naturally picks up what, then encode that. Routing rules written before you have seen the pattern usually describe how you imagined the team works.
The metrics worth arguing about
Two numbers get watched too much and one gets watched too little.
| Metric | Watch it because | Careful of |
|---|---|---|
| First response time | It is what the customer feels | An average hides the worst cases; look at the slowest tenth |
| Tickets resolved per agent | Rough capacity signal | Rewarding it produces fast closes and high reopens |
| Reopen rate | It is the only honest test of resolution | Almost nobody tracks it |
If you take one thing from this guide, make it reopen rate. A team that closes fast and reopens often is not performing well, it is pushing work into next week. A rising reopen rate against flat resolution time is the earliest warning sign a support operation gives you.
Read a handful of reopened tickets each week and the cause is usually obvious within minutes: an answer that was correct but not actionable, or a resolution recorded before the customer confirmed anything.
Do not compare your numbers to published benchmarks. Support metrics vary enormously by industry, price point and channel, and a figure from an American software company tells an Indian jewellery brand nothing useful. Compare this month with last month, for the same team, and improve from there.
Example conversation
Run support like a desk, not an inbox
Start free in minutes — no code, no credit card. Bring your own AI keys and pay no per-message markup.
Frequently asked questions
Do we need ticketing if we only use WhatsApp?
Channel count is not the trigger, volume is. One busy WhatsApp number with thirty conversations a day and three people answering already has the coordination problem that ticketing solves.
Will customers know they are in a ticketing system?
No, unless you mention a ticket number. They keep chatting on WhatsApp exactly as before. Everything the desk adds is on your side of the conversation.
What happens to tickets when an agent leaves?
Reassign them in bulk from the worklist. Because ownership is recorded rather than remembered, nothing is lost when somebody goes, which is the main practical argument for the change.
Is the Support Desk included in the plan?
It is a paid add-on at ₹2,999 a month, switched on per workspace and billed on top of your plan. The ordinary shared inbox and live chat are included in every plan.
Can the AI keep answering while tickets exist?
Yes. The bot handles what it can and the ticket records it, and anything handed over lands in your team's worklist. Most teams find the bot closes the majority of tickets without a human.
How long does it take to set up?
About twenty minutes for working hours, deadlines and a short tag list. The tickets themselves need no setup at all, because they open automatically from existing channels.
Related reading
Available across India — see WhatsApp & AI calling automation by state and city.