What was said
What this says
As at Fri 9 Oct 2026Laura Penhaligon explained to Hugh Fenton and Toby Cowell how ticket and invoicing data flow through a Snowflake staging layer. She also set the nightly refresh expectation of seven in the morning. The customer was engaged and raised no concerns.
What happened: On 11 Mar 2026 Laura walked Hugh and Toby through the build. A connector brings ticketing data into a staging layer in Snowflake, and the brain queries the cleaned data.
Customer view: Hugh asked what happens if data looks wrong and what BG-Demo needed from them that week. He accepted the answers and agreed a follow-up call at the same time the next week.
Commercial: Laura told them the nightly refresh completes by seven in the morning. That sets a freshness expectation from the first call, which matters for the renewal.
Watch: Connector sign-off depends on Toby reviewing a sample of about a hundred records, roughly forty minutes per connector. Toby is the hands-on contact for that work, so his availability drives the pace.
Themes
As at Fri 9 Oct 2026Who was there
BG-Demo
Them
Follow-up
Next steps
As at Fri 9 Oct 2026- Laura sends the ticket sample data by Thursday before the next call.
- Laura and Toby review the sample together and sign off the ticket connector, then repeat for invoicing.
Risks
As at Fri 9 Oct 2026- The nightly seven in the morning refresh was stated as a promise, so any slip will be noticed early.
- Connector sign-off relies on Toby finding time to review sample data each week.
Opportunities
As at Fri 9 Oct 2026- A clean ticket and invoicing sign-off builds confidence for the remaining connectors.
Transcript
- Laura PenhaligonToby, good to see you. How's the week been? Did you manage to get the Snowflake test account set up?
- Toby CowellYeah, did it yesterday. Got the credentials sorted, you should be able to log in.
- Laura PenhaligonGreat. Let me jump into your environment quickly and check the schema... okay, I can see your ticket system tables, invoicing tables are there. Good structure. Now I want to walk you through what we're building this week. I'm going to create a connector from your ticketing system into a staging layer in Snowflake, then the brain will sit on top of that.
- Toby CowellSo the brain pulls from the staging layer?
- Laura PenhaligonExactly. To be precise, we bring your raw data in, we normalise it: clean up any duplication, standardise date formats, that sort of thing, then the brain queries the clean data. Means if something goes wrong, we fix it in the staging layer, the brain sees the fix automatically.
- Hugh FentonWhat if we find an issue, data that looks wrong?
- Laura PenhaligonYou'd message me or the helpdesk, we'd dig into the source system, see what's happening. Often it's just that your system is recording something differently than we expected. We adjust the mapping and it's fixed on the next refresh.
- Toby CowellWhen does the refresh happen?
- Laura PenhaligonNightly, complete by seven in the morning. So if a ticket closes at five in the evening, it'll be in the brain by the next morning.
- Hugh FentonThat's good. Okay, what do you need from us this week?
- Laura PenhaligonI'll pull a sample of your ticket data, maybe a hundred records, and we'll review it together. I want to check that what I'm seeing maps to what you expect. Then we can sign off on the connector. I'll do the same with invoicing.
- Toby CowellHow long does that take?
- Laura PenhaligonReview call is probably forty minutes per connector. I'll send you the sample data before we meet, you can have a look, then we'll chat through any oddities.
- Hugh FentonSounds good. Same time next week?
- Laura PenhaligonPerfect. I'll send the samples by Thursday. Let me put the next call in the diary. Cheers, both.