Customer support · Production

Wave Financial achieves 5x ROI and $1.2M annual savings using Ada's chatbot Mave

The problem

Wave Financial's customer support team faced seasonal Q1 spikes of 200-300% in inquiry volume. The company managed these with an unsustainable 'all hands on deck' approach—pulling in staff from other departments and having agents work overtime—leading to longer wait times, negative customer interactions, and missed revenue opportunities.

First attempt

Wave's previous approach of deploying all-hands support during peak season—using staff borrowed from other departments and mandatory overtime—was explicitly described as unsustainable as the company grew.

Workflow diagram · grounded in source
1
Customer inquiry triggers bot
Trigger
Customer inquiries arrive through Wave's support channels, including during seasonal Q1 volume spikes.
source quote
“Wave's Customer Experience team ramps up for a mega spike in customer support volumes—upwards of 200-300%”
2
Mave handles FAQs 24/7
Ai action
Mave, Wave's chatbot built on Ada, answers redundant FAQs about products and features around the clock.
source quote
“Mave works 24/7, answering redundant FAQs such as basic information about products and features”
3
Tier- and region-based routing
Routing
Ada intelligently routes customers to different support levels and teams based on their account tier and region.
source quote
“Katie and her team are using ada to intelligently route different customers, based on their account tier and region, to different levels of support and different teams”
4
Complex inquiries to human agents
Human review
Conversations requiring a sophisticated human touch are passed to Wave's agents for resolution.
source quote
“the conversations that are left for Wave's agents are more complex and need a sophisticated human touch”
Reported outcome

Wave achieved a 5x return on investment within 12 months, with $1.20 million in estimated savings from inquiry deflections, a 65% reduction in year-over-year support ticket creation within the first month, and a 70% containment rate during the busy season at launch.
Over 500 million interactions have been automated in total.

Reported metrics
Year-over-year support ticket creation reduction (first month)65%
Interaction containment rate during busy season at launch70%
inquiry deflection savings (March 2020 – March 2021)$1.20 million
Q1 customer support volume spikeupwards of 200-300%
Show all 5 reported metrics
year-over-year support ticket creation reduction (first month)65%
interaction containment rate during busy season at launch70%
inquiry deflection savings (March 2020 – March 2021)$1.20 million
Q1 customer support volume spikeupwards of 200-300%
time to launch chatbotthree and a half weeks
Reported stack
AdaMaveEngage
◆ 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/wave
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Wave achieved a 5x return on investment within 12 months, with $1.20 million in estimated savings from inquiry deflections, a 65% reduction in year-over-year support ticket creation within the first month, and a 70% c…

What tools did this team use?

Ada, Mave, Engage.

What results were reported?

Year-over-year support ticket creation reduction (first month): 65%; Interaction containment rate during busy season at launch: 70%; inquiry deflection savings (March 2020 – March 2021): $1.20 million; Q1 customer support volume spike: upwards of 200-300% (source-reported, not independently verified).

What failed first in this deployment?

Wave's previous approach of deploying all-hands support during peak season—using staff borrowed from other departments and mandatory overtime—was explicitly described as unsustainable as the company grew.

How is this customer support AI workflow structured?

Customer inquiry triggers bot → Mave handles FAQs 24/7 → Tier- and region-based routing → Complex inquiries to human agents.

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.