What it solved.
What it couldn’t.

Insights show how your AI customer service agent is doing. See how many conversations Dazza handled on its own, why it handed over, how customers rated their chats and the questions it couldn’t answer. Then you know exactly what to improve.

EVERY CONVERSATION → THE BIG PICTURE30 DAYS
BETTER WITH EVERY INSIGHT.
78% HANDLED WITHOUT A PERSON (EXAMPLE)
Why RECORDED FOR EVERY HANDOVER
CSAT CUSTOMERS RATE EACH CHAT AS IT ENDS
£ USAGE AND COST, AS IT HAPPENS

Fewer charts.
Better questions.

Four views that tell you how it’s going
and what to do next.

01 / RESOLUTION

How much it handled.

Chats by day, and how many Dazza answered fully without bringing in your team.

02 / HANDOVERS

Why people stepped in.

Every handover keeps its reason: asked for a person, couldn’t answer, a complaint, or something only staff can do.

03 / GAPS

What it couldn’t answer.

The questions that stumped it, each linked to its conversation. Add the answer and the gap closes from the next one.

04 / USAGE

What it costs, as it happens.

Customer replies, preview replies and your monthly total, so the bill is never a surprise.

Every gap it finds
is an answer you only
write once.

Insights lists what Dazza couldn’t answer. Add the answer in your own words and it’s used from the next conversation on, in chat and in email.

All in one page.
No setup.

Pick the last 7, 30 or 90 days. Everything updates as conversations happen.

01

Conversations per day

Every chat, by the day it started, so you can see the busy days.

02

Handled by Dazza

Chats Dazza answered fully, without a person.

03

Handed to the team

How many needed you, and the reasons why.

04

Customer rating

The share of chats rated Good when they ended, beside the thumbs on single answers.

05

Workflows

How often each one ran, and how it finished.

06

Couldn’t answer

Questions with no answer in your knowledge, newest first.

07

Topics

What customers ask about, labelled for you, with how often Dazza handled each.

08

How customers felt

Happy, neutral or unhappy, read from every conversation.

09

By channel

Web chat, email, WhatsApp and the rest, side by side, or one at a time.

10

Weekly report

Every Monday morning, an email from Dazza: top questions, what it handled, answers to approve and one thing to do.

Insights,
explained.

More in the help centre.

What counts as handled?

A chat where Dazza answered every question without handing it to your team. Chats it handed over, or couldn’t answer, are counted separately so the figure stays honest.

How do customers rate their chats?

When a chat on your website or in your app ends, Dazza asks “How was this chat?”: Good, Okay or Poor, with an optional comment. Insights shows the share rated Good, with a link to the chats rated Poor. Each of Dazza’s replies also has a thumbs up and a thumbs down, shown as answer ratings. See Customer ratings.

Do preview chats skew the figures?

No. Preview and test conversations are kept out of your figures and never count towards your plan.

Do the figures include email and messaging?

Yes. The figures cover every channel: web chat, your app, email and messaging apps. Pick a channel to see it on its own, or compare them under By channel.

Can we get the figures by email?

Yes. Every Monday morning Dazza emails a weekly report: what customers asked most, the share it handled, the questions it couldn’t answer with suggested answers to approve, ratings and one recommendation. Admins get it by default; anyone can turn it on or off. See The weekly report.

Where do topics come from?

Dazza labels each conversation with a few topics, how the customer felt and their language. It uses your own list of topics when you set one, and you can relabel any conversation in the inbox.

YOUR AI CUSTOMER SERVICE AGENT

Less back-and-forth.
More sorted.

A WALL & FIFTH PRODUCTGOOD SERVICE. GOOD FEELING.