Renewal rate, GermanySample data
- Yesterday
- 74.6%
- Usual
- 91.2%
- Declined
- 212 of 1,277
- At risk this week
- €184,000
German renewals fell
212 invoices went out without VAT numbers after Friday’s billing change.
Alpenglow
MenuAI analytics that reads every metric overnight.
Request accessConcept project by al. Fictional product and data.
Alpenglow is the light that touches the highest peaks first. Your morning should work the same way.
Sample workspace, Tuesday at 06:40. Three findings out of 2,140 metrics, each with its cause and its size.
212 invoices went out without VAT numbers after Friday’s billing change.
Yesterday’s one-page checkout added 6.4 points. Android has not changed.
A vendor API change broke contact sync for 38 accounts.
They moved inside their usual range, so they get no alert and no paragraph.
| Metrics | Source | Count | Read between |
|---|---|---|---|
| Money, 612 metrics | |||
| Renewals | Billing | 148 | 00:00–00:14 |
| New contracts | Billing | 96 | 00:14–00:23 |
| Refunds and credits | Billing | 84 | 00:23–00:31 |
| Invoices and dunning | Billing | 132 | 00:31–00:43 |
| Revenue by plan | Warehouse | 152 | 00:43–00:57 |
| Product, 884 metrics | |||
| Sign-up and activation | Events | 214 | 00:57–01:08 |
| Checkout | Events | 178 | 01:08–01:17 |
| Feature use | Events | 336 | 01:17–01:34 |
| Speed and errors | Logs | 156 | 01:34–01:42 |
| Support, 236 metrics | |||
| Tickets by integration | Help desk | 124 | 01:42–01:49 |
| Time to first reply | Help desk | 72 | 01:49–01:53 |
| Satisfaction | Surveys | 40 | 01:53–01:56 |
| Marketing, 408 metrics | |||
| Paid channels | Ad accounts | 244 | 01:56–02:05 |
| Organic and referral | Analytics | 164 | 02:05–02:10 |
Per workspace, billed monthly. Sample figures.
No. Alpenglow runs its queries inside your warehouse through a read-only role. The model receives aggregates and metric names, never customer rows.
Three is the limit we set, so each one gets a cause, a size and an owner. The rest wait in an appendix you can open.
The reading says so in one line. Some mornings have no findings, and those are good mornings.
Each metric gets its own seasonal baseline from at least 13 months of history. You can pin known events, such as a launch or a price change.
Yes. Every finding links to the SQL it ran and the metrics it ruled out, so your analysts can repeat it.
A dashboard lights everything and waits for you to look. Alpenglow does the looking and shows you what changed.
Your first reading arrives within a week of connecting.