Spark saves 2 minutes a call with AI assistants
In plain words
Spark's 350+ frontline advisors ask AI assistants instead of hunting for answers. That's around 20,000 questions a month.
What happened: About 2 minutes saved per call.
Source: news.microsoft.com ↗
How to do it yourself
Our suggested steps. The company didn't publish theirs.
- 1
Record calls (with consent) or keep rough notes in your CRM.
- 2
After each call, paste the transcript/notes into Copilot or ChatGPT with the prompt below.
- 3
Paste the summary and next steps into the CRM record.
- 4
Use the same prompt to search 'what did we promise this customer last time?'.
- 5
Time it for a week: before vs after.
The instructions to give the AI
Paste this into ChatGPT (the free plan is fine), swap the [brackets] for your details, and test it on a few real examples.
Agent instructions (prompt)
Summarise this customer call for our CRM. Output: Reason for call | What we did | Promises made (with dates) | Follow-ups (owner, due) | Customer mood (1-5) | Upsell/risk flags. Max 80 words. Transcript/notes: [paste]
[Highlighted] = replace with your own details before you run it.
What you should get back: "Reason: late delivery | Did: re-sent order | Promised: refund by Fri | Follow-up: Sam, Thu | Mood: annoyed→calm"
Written by Quickly and Good, inspired by this case.
More detail: tools, prices and what you'll need
The company used: Microsoft Copilot
You could use: Microsoft 365 Copilot (or ChatGPT) for call summaries + your CRM
- Microsoft Copilot
AI assistant and agent builder inside Microsoft 365 and Teams.
See pricing · Official site ↗
- ChatGPT
General AI assistant: writes, summarises and answers from files you give it.
Free plan; paid plans available · Official site ↗
Prices from each tool's own pricing page, checked Oct 2026.
What you'll need
- ☐A ChatGPT or Claude account
- ☐Your own documents, templates or examples
- ☐The agent instructions (prompt)
- ☐A habit of checking every output before you use it
Can I do this myself?
Yes. You can try this today with a chat assistant and your own files.
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.