1-800-Flowers tested AI agents on chat and phone, and 93% of customers were happy
In plain words
1-800-Flowers.com tested Decagon on chat and voice, where emotional, time-critical orders are common.
Still a trial: About 93% of customers rated it well, well above its old automated system.
Source: decagon.ai ↗
How to do it yourself
Our suggested steps. The company didn't publish theirs.
- 1
Pick your most time-sensitive order type (e.g. same-day) and pilot on that.
- 2
Rewrite your help articles for those topics so each has one clear answer.
- 3
Connect Decagon AI agents to your helpdesk and knowledge base and paste the prompt below as its instructions.
- 4
Run it on 50 real past tickets and fix every wrong or vague answer.
- 5
Go live on one channel, review a sample every week, then add channels.
The instructions to give the AI
Paste this into Decagon (the free plan is fine), swap the [brackets] for your details, and test it on a few real examples.
This is one part of the setup. It won't connect to your phone, inbox or other apps on its own; the steps above cover that.
Agent instructions (prompt)
You help [store] customers with orders that often matter a lot (birthdays, funerals). Check the order status and delivery window first. Be warm. If a delivery will miss an event, offer [remedy] and hand to a human for anything beyond it.
[Highlighted] = replace with your own details before you run it.
What you should get back: "Your flowers are out for delivery and will arrive by 5pm. If they miss the service, we'll redeliver free."
Written by Quickly and Good, inspired by this case.
More detail: tools, prices and what you'll need
The company used: Decagon AI agents
You could use: Decagon AI agents
- Decagon
Enterprise AI support agents for chat, email and voice.
See pricing (sales-led) · Official site ↗
Prices from each tool's own pricing page, checked Oct 2026.
What you'll need
- ☐An enterprise AI platform licence
- ☐Integration with your core systems (CRM, billing, identity)
- ☐Security, privacy and compliance review
- ☐A curated knowledge base
- ☐A testing and evaluation programme
- ☐Human escalation and monitoring
Can I do this myself?
Not on your own. These are enterprise platforms sold with implementation help. A small business can get part of the way with Intercom Fin or Tidio Lyro (3-7 days, no-code), but matching this result means hiring a builder.
Similar cases
CMA CGM's AI voice agents resolve 46% of cargo release calls
More than 46% of cargo release enquiries are resolved by the AI agents, and about a third of 4,000 customers called by the sales agent moved on to active discussions.
ElevenLabs' own AI support agent resolves 65% of its tickets
65% of tickets resolved end to end, first reply in about 60 seconds instead of hours, and 70% positive ratings.
Telstra lets an AI agent sort refunds while customers wait for repairs
Refunds are worked out and paid automatically once service is back, which takes admin off 1,000+ staff.