← Back to Knowledge Quest For contact centres

Read every interaction. See what dashboards miss.

Voice, chat, messaging, email - Knowledge Quest reads every transcript in your contact centre, finds what's driving sentiment down and cost up, and drafts the articles that close the gaps.

Book a demo → See a sample brief

90% → 55%

CSAT collapse when customers need 3+ contacts

- COPC Inc.

$13.50

average assisted contact cost vs $1.84 for self-service

- Gartner, 2024

Only 5%

of contact centres hit world-class FCR (80%+)

- COPC Inc.

Why dashboards miss it

Your CCaaS shows you what happened. It can't tell you why.

Real-time dashboards surface volume, AHT, and queue depth. But when CSAT drops on Tuesday or resolution times stretch on a Thursday, your analytics platform can't read the text of those interactions - it can only count them.

Your dashboards say

CSAT dropped 8 points.

You can see what, but not why. Is it one queue? One topic? One day? One agent? Every investigation takes hours.

Knowledge Quest says

This is why.

"Billing queue. Double-charge refund calls up 42% after last week's batch processing error. 94% of agents are escalating because no article covers the new refund flow. Draft ready."

What you couldn't see

The trajectory.

The 8-point drop is week 6 of a slide that started when a process changed. Without understanding the underlying interactions, no one connected the dots.

What it finds

Signal, not noise - across every channel.

Knowledge Quest reads every interaction your contact centre handles - voice inbound, voice outbound, chat, messaging, email - and turns the text into actionable intelligence on a cadence your team already lives by.

Every interaction, every channel

Voice transcripts, chat sessions, messaging threads, email conversations. Not just structured fields - the actual text of what was said.

Voice Chat Messaging Email

Sentiment by queue, topic, taxonomy

Not just an overall score - sentiment broken down by the dimensions that actually matter. See which queue is burning trust and which article would fix it.

Emerging topics, early

Spikes detected before they show up as volume. A novel topic with 12 interactions this week is already on the brief - before it becomes a crisis next week.

Callback & re-contact tracking

Which articles actually resolve first-contact? Which ones trigger a callback? Attribution per article per topic - closed-loop, not guessing.

Gap detection from real intents

The interactions you handle every day tell you what your KB doesn't cover. We rank those gaps by dollar cost, not "popularity".

Copilot-drafted articles & scripts

Agent scripts, KB articles, desktop guidance - drafted from real interaction evidence. Every sentence traces back to its source transcript.

Handle time, decomposed

True voice AHT (native handle time) broken into talk, hold, after-call wrap-up and connect - so you see where the minutes actually go, by queue and topic.

AHT Talk Hold Wrap-up

On every interaction: measured, wrapped up, accounted for.

Beyond the brief, Knowledge Quest works call by call - checking the behaviours you care about, writing the wrap-up, and treating email and other async work on its own terms.

Tracked measures

Describe a behaviour in plain English - "retention offer made", "offer accepted", a compliance step - and every new interaction is assessed for it. Kai can help draft the definition; test it on up to 10 recent interactions, then switch it on. Results appear in Kai (including by agent, with drill-down to the calls), interaction search, the interaction detail view and your Insight briefs, which stay at team level with no agent names.

After-call notes

A structured wrap-up record for your CRM, written from the transcript: reason for contact, what was discussed, actions, commitments and follow-ups. Narrative or bullets; copy or download. Numbers and identifiers that speech recognition may have misheard are flagged "Verify before filing", with the original wording.

Email & async channels

Email and other async interactions are ingested as they arrive and refreshed daily while they're open. Work time (hands-on) and response time (waiting in queue) are reported separately per channel, with a backlog "as of" figure and true handle time - in dashboards, briefs and Kai.

Work time Response time Backlog

Interaction detail Illustrative · sample data

Voice · Cancellations queue · sample call

Tracked measures

Retention offer made✓ Met

Offer accepted✗ Not met

Identity verified before changes✓ Met

After-call note

Reason for contact: wants to cancel after a price change.
Discussed: retention offer and alternative plans.
Commitment: agent to email a plan comparison today.
Follow-up: customer to confirm their choice.

Verify before filing · Reference "4815" - original wording: "four eight one five".

Ask Kai

Ask why the queue moved. See the calls behind it.

Kai is the grounded analyst inside Knowledge Quest. Kai finds and explains the gap; Knowledge Copilot drafts the fix.

Start from a queue, a chart, an alert or a brief - Kai opens already scoped to what you're looking at.

Every answer cites the actual calls and articles as clickable chips, and "How Kai got this" shows the working behind every figure.
Live charts and tables, grouped by queue, channel, topic or agent - looking back a full 12 months.
Tracked-measure results by agent, with drill-down to the calls behind each count.
Charts export to CSV; whole conversations to a branded PDF or text.

See Kai → Book a demo

Kai analysis

Trust & rigour

Built for accountable contact centre operations.

Knowledge Quest is designed for production contact centre environments: deterministic where operational accuracy is critical, controlled use of language models where interpretation adds value, and full traceability for review and governance.

Deterministic where it matters

Core scoring, alerts, and operational metrics are rule-based and auditable.

AI with guardrails, humans in control

Language models handle bounded drafting and narrative tasks under structured controls; threshold-gated workflows keep your team in control of ambiguous decisions.

Every number checked

Every number in a brief is checked against your interaction data and corrected before it's sent. Briefs only write sections for data you actually have, and know your holidays and quiet periods.

Full provenance

Recommendations and outcomes are traceable end-to-end for governance and review.

Enterprise SSO

Data residency

Runs in available Microsoft Azure regions, with optional self-hosted AI inside your own subscription.

Your stack

Sits next to your CCaaS. Reads every interaction.

Knowledge Quest connects to your existing contact centre platform and knowledge base - no rip-and-replace. We read transcripts and interaction data; you keep your tools.