Customer support · Production

Epos Now saves 60,000 human labor hours per month and achieves 70% automated resolution rate with Ada's AI agent Sidekick

The problem

Epos Now's B2B customer support was strained by complexity across dozens of verticals, with customers demanding high-touch, multi-step guidance that legacy IVR and scripted chatbots could not deliver. Support channels operated in silos with different technologies and no unified knowledge base, producing inconsistent customer experiences.

First attempt

Epos Now's IVR system was pre-configured around scripted routing and failed to deliver the quality of experience they wanted, with customers sometimes ending up with the wrong agent and experiencing longer wait times. Scripting responses across a diverse customer base quickly led to diminishing returns.

Workflow diagram · grounded in source
1
Customer inquiry arrives
Trigger
Customers reach out via messaging or phone channels and Sidekick handles complex inquiries in real time.
▾ source quote
“Epos Now launched their AI agent, called Sidekick, on messaging to handle complex inquiries in real time, delivering instant, seamless support”
2
Knowledge base retrieval
Ai action
Sidekick pulls from Epos Now's extensive knowledge base and system integrations to provide customers with accurate, up-to-date information.
▾ source quote
“Sidekick pulls from various sources, including Epos Now's extensive knowledge base and system integrations, to provide customers with accurate, up-to-date information”
3
AI-guided resolution
Ai action
Sidekick guides customers through complex processes with the same precision and personalization that once required human intervention.
▾ source quote
“Epos Now guides customers through complex processes with the same precision and personalization that once required human intervention”
4
Pre-handoff case documentation
Output
When Sidekick cannot fully resolve an inquiry, it creates a case, documents the interaction, and takes notes before the customer is handed off.
▾ source quote
“like creating a case, documenting the interaction, and taking notes—has already been done by the time the customer is handed off”
5
Human agent review
Human review
Human agents receive a fully captured and organized case and need only update it with the resolution.
▾ source quote
“When it reaches a human agent, everything is already captured and organized, leaving them to simply update the case with the resolution”
6
Team monitors and refines AI
Feedback loop
The support team monitors AI conversations, refines customer workflows, and ensures Sidekick stays continuously aligned with product updates.
▾ source quote
“the team specialized in monitoring AI conversations, refining customer workflows, and ensuring that Sidekick was continuously aligned with product updates”
Reported outcome

Sidekick now automates 70% of support demand across all channels, saves 60,000 human labor hours per month, increased CSAT by 30% on messaging and 15% on voice, and call demand dropped by 22% to 25%, while customer service teams delivered 40% year-over-year revenue growth.

Reported metrics
Human labor hours saved per month60,000
CSAT increase on messaging30%
CSAT increase on phone15%
Automated resolution rate (combined channels)70%
Show all 8 reported metrics
human labor hours saved per month60,000
CSAT increase on messaging30%
CSAT increase on phone15%
automated resolution rate (combined channels)70%
revenue generation growth YoY by customer service teams40%
call demand reduction22% to 25%
phone inquiries auto-resolved by Sidekick40%
average automated resolution rate60% to 70%
Reported stack
AdaSidekickIVR
◆ Does this fit your context?

Compare to your context

Tell us your scale, team, and constraints. We'll show what changes at your size, what fails at your scale, and whether this case is a fit, needs adaptation, or won't scale to you. Free demo, no signup.

Compare to your context →
~30 seconds · free
Source
https://www.ada.cx/case-study/epos-now
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Sidekick now automates 70% of support demand across all channels, saves 60,000 human labor hours per month, increased CSAT by 30% on messaging and 15% on voice, and call demand dropped by 22% to 25%, while customer se…

What tools did this team use?

Ada, Sidekick, IVR.

What results were reported?

Human labor hours saved per month: 60,000; CSAT increase on messaging: 30%; CSAT increase on phone: 15%; Automated resolution rate (combined channels): 70% (source-reported, not independently verified).

What failed first in this deployment?

Epos Now's IVR system was pre-configured around scripted routing and failed to deliver the quality of experience they wanted, with customers sometimes ending up with the wrong agent and experiencing longer wait times.

How is this customer support AI workflow structured?

Customer inquiry arrives → Knowledge base retrieval → AI-guided resolution → Pre-handoff case documentation → Human agent review → Team monitors and refines AI.

Related customer support cases

WHAT TO DO WITH THIS

Now compare it to your context

This case is one data point. Whether its pattern fits you depends on your volumes, your stack, and your exception load — that comparison is the step no case study can do for you.

Browse the Customer support cases →