Gupshup chatbot analytics
Read what your Gupshup agent actually did for each customer, scored against your own rules, beside the delivery counts Gupshup already gives you.
Gupshup counts messages well. Totals both ways, conversations, unique users, returning users, and the averages across them.
Campaign reporting counts delivery: triggered, sent, delivered, read, failed. Goal analytics counts whether a session walked every milestone you defined.
Every one of those is countable from the message envelope. None of them reads what the agent said.
The gap that leaves
Two bots with identical message counts, identical delivery rates and identical goal completion can differ completely in whether customers got helped.
| Gupshup reports | How many messages, to how many people, and whether they arrived. |
|---|---|
| We report | Whether the reply answered the question, against rules you approved. |
| Together | Volume you can trust, and quality you can act on. |
What gets scored
- Whether the agent answered what the customer asked.
- Whether it admitted it was not a person when asked.
- Whether it invented a price or a delivery date.
- Where the customer gave up, and the reason.
Hindi, Tamil, Telugu, Hinglish and English are scored by the same rules, which matters on Gupshup more than on most platforms in this list.
Connect it
Set the Evidova URL as your Gupshup callback.
We read the messages Gupshup already posts to your webhook.
Import the history you want a baseline from.
The same rules run over it, so the first chart has a past in it.
Read next
- Quickstart: connect a platform · Ten minutes, one path, from a webhook to your first score.
- Webhook reference · The URL, the signature, and what we answer a delivery with.
- All twelve platforms · Every voice and chat stack Evidova reads, side by side.