Skip to main content
Freshworks

On Freshworks - Freddy AI Activation

You're licensed for Freddy AI. Now get the value.

A fixed-scope activation of the Freddy AI capabilities your agreement actually entitles you to, measured against a pre-engagement baseline, with a value report your CFO can take to renewal. The same measured-activation discipline we apply to agentic AI on enterprise platforms.

Week 1

Baseline agreed and captured with your sponsor

Measured

Deflection, resolution time and agent productivity vs baseline

30 days

Post-handover advisory included

The four capabilities

Freddy AI is four products. They are not activated the same way.

Most activation projects stall because "switch on Freddy AI" is not one job. Each capability has a different owner, a different readiness test and a different licensing position. Pick the one you are being asked about.

Select a Freddy AI capability

Bundled allowance on Enterprise, capped by session volume

Autonomous resolution on chat, messaging and email. This is the capability with a number attached, because deflected requests are countable, so it's usually where the business case starts.

What we activate

Sequenced by demand, highest volume first.

  • Intent coverage built from your actual top request types, not a generic library
  • Knowledge grounding so answers come from your articles rather than improvisation
  • Escalation design deciding when the agent stops and a person takes over
  • Deflection measurement against the pre-activation baseline we capture first

What it needs from you

Checked in week one.

  • Knowledge articles that are current
  • Clean request categorisation
  • Agreed escalation rules
  • A named service owner
  • Session volume headroom

Licensing positions change between Freshworks releases. We confirm what your agreement actually entitles you to in week one, before anything is scoped, because discovering a capability is an add-on halfway through is how activation projects lose their sponsor.

The renewal question

The AI is switched on. The value story isn't.

Freddy AI is not one switch. Agents resolve requests, Copilot assists your service desk, Insights reports to your leadership, and Agent Studio builds the rest. Most organisations have some of it enabled, none of it measured, and a knowledge base that quietly limits what any of it can do.

Enabling a feature is not activating value. Activation means the capabilities are configured to your workflows, grounded in knowledge that's fit to ground them, adopted by your agents, and measured, so the value survives contact with a renewal negotiation.

  • Freddy capabilities enabled ad hoc, with no baseline and no measurement.
  • Knowledge gaps and stale articles quietly cap both resolution rates and assistant quality.
  • Renewal approaches with "we think it helps" as the entire value story.

Four committed outcomes

Live, grounded, measured, and yours to run

  • Operational

    Agreed Freddy AI capabilities live in production: configured to your workflows and risk appetite, not just toggled on.

  • Grounded

    Knowledge coverage assessed and remediated where it matters most, because AI answers are only as good as the knowledge beneath them.

  • Measured

    Deflection, resolution time and agent productivity tracked against the pre-engagement baseline your sponsor signed off.

  • Defensible

    A value report with explicit methodology: evidence for the renewal conversation, branded as yours, not consultancy marketing.

How it runs

Baseline to value report, in gated phases

  1. Phase 1

    Baseline and readiness

    Platform review, knowledge audit, baseline metrics agreed and captured.

  2. Phase 2

    Configure and ground

    Capabilities configured, knowledge remediated, piloted with a real agent group.

  3. Phase 3

    Activate and measure

    Production rollout, adoption support, measurement against baseline begins.

  4. Tail

    Advisory window

    Thirty days of post-handover advisory while your team takes over.

If the knowledge foundation cannot carry deflection yet, we say so at gate one, and the remediation becomes part of the plan, not a surprise after go-live.

Who it's for

For teams that need the AI to show up in the numbers

  • Existing Freshservice customers
  • Heads of Service, Service Desk Managers, IT Directors
  • Renewal or expansion decision ahead

Measured, not asserted

Measurement is designed in from week one against a baseline your sponsor signs off, never reverse-engineered later.

Grounding before features

Knowledge and data readiness come first; capabilities are activated in the order the foundation can support.

No managed-service capture

Your team runs it when we leave: knowledge transfer and a thirty-day advisory tail are part of the scope.

Make the licence earn its renewal. Let's talk.

Tell us when your renewal lands and we will work backwards from it.