KPI audit: send one plant, one month.
Once, on one plant, so you can judge the work. We recompute performance ratio and availability from your raw readings, with the rules written down. Then we show how far each portal's number is from yours, which choices explain the gap, and what stays open.
You send
- Portal access or a CSV export for one plant and one month
- The master data: capacity, inverters, sensors
- Anything that changed in the month
You get
- Performance ratio and availability as a base row, with the rules written down
- The difference against each reported number, with what explains it and what stays open
- A method card next to every number
- A 30-minute walkthrough
Nothing changes in your monitoring. No migration.
Start with one plant.
Four fields. You get the list of what to send by email right away.
What the report looks like.
One page of the audit report, filled with the same synthetic plant as our examples.
Audit report: one plant, one month
One page of the report, October 2026
Performance ratio: base row and what explains the difference
| Layer | Source A | Source B | Source C |
|---|---|---|---|
| your rules, written down | 79.0 % | 79.0 % | 79.0 % |
| Energy out | +0.8 | −0.4 | ±0.0 |
| Sunlight | +1.6 | −2.0 | −0.4 |
| Size | ±0.0 | −0.3 | ±0.0 |
| Hours | +0.5 | ±0.0 | −0.2 |
| not reproduced | ±0.0 | −0.2 | ±0.0 |
| reported | 81.9 % | 76.1 % | 78.4 % |
Source B's remaining −0.2 could not be reproduced. It stays open.
Availability: reported number against your base row
| Source | reported | base row | what moves it |
|---|---|---|---|
| Source A | 99.5 % | 97.7 % | 24 hours a day, inverters counted, gaps count as available, grid outages excluded |
| Source B | 97.5 % | 97.7 % | gaps count as down |
| Source C | 99.2 % | 97.7 % | contract exclusions: grid outage, maintenance within quota |
The method, written next to the number
Technical availability
Synthetic example. The numbers are made up, the mechanics are not.
What happens after the report.
Step 1: the report
We do it once, on one plant.
You send one plant and one month. We recompute the KPIs and show you what we find, so you can judge the work before deciding anything else.
- One plant, one month
- Nothing to install
- Nothing changes in your monitoring
Step 2: every plant, continuously
Then it runs all the time.
The same definitions on every plant, new data as it arrives, checks every night. We run it for you:
One machine, just for you
It runs on a machine in Germany that you share with no one else. We operate it.
Your data, in your storage
Raw data and analytics stay in your own highly available storage. You can revoke our access at any time.
Your plant changes, in your project
Your team can edit master data and see who changed what.
If the machine fails, nothing is lost. It is rebuilt from your storage and your project.
Running it continuously costs one yearly subscription, machine included. You pay your storage provider directly.
Make the base row yours.
| Layer | Source A | Source B | Source C |
|---|---|---|---|
| your rules, written down | 79.0 % | 79.0 % | 79.0 % |
| Energy out | +0.8 | −0.4 | ±0.0 |
| Sunlight | +1.6 | −2.0 | −0.4 |
| Size | ±0.0 | −0.3 | ±0.0 |
| Hours | +0.5 | ±0.0 | −0.2 |
| not reproduced | ±0.0 | −0.2 | ±0.0 |
| reported | 81.9 % | 76.1 % | 78.4 % |
Synthetic example. The numbers are made up, the mechanics are not.
Audit
The KPIs your money depends on, recomputed by you from the raw readings. In this example, A's 81.9 comes from +0.8 for the inverter sum, +1.6 for one named sensor and +0.5 for dropping hours with one side missing.
Compare
One base row for every plant, whichever portal it runs on. Twelve assets, three providers, one number that means the same thing.
Let the portals monitor. They are good at it.
What stays true while we do it
Raw data stays as it arrived. Every number carries its origin. Same inputs give the same numbers.
You talk to the founder.
Philipp Wegner, fifteen years in solar monitoring and data engineering, previously on systems taking care of over 50,000 plants, currently monitoring over 500 MW. You speak directly with the person who built the platform and knows these numbers from the inside.
Questions
Is the report a one-off?
Yes. It is one plant and one month, done once, so you can judge the work. Nothing continues unless you decide it should.
What if we want this on all our plants?
Then it runs continuously: every plant, new data as it arrives, checks every night. We run it on one machine in Germany, just for you.
Where does our data live?
In your own highly available storage, separate from the machine. It has to be S3-compatible, and we can set it up for you. It is your account, so it is also your responsibility, and you can revoke our access at any time.
Can we check your numbers ourselves?
Yes. You have the raw data and the rules, so you can recompute any number yourself, and the result has to match.
Can we use the data without you?
Yes. It is stored in an open format, data lake ready, and you can query it directly with common data engines.
What happens if the machine fails?
Nothing is lost. Your raw data and analytics are in your storage. Changes to your plants, such as master data, are in your own project, where your team can edit them and see who changed what. Every manual correction is recorded in one of the two, with the reason and a name. Everything is rebuilt from those two on a new machine.
What do you need from us?
Portal access or a CSV export for one plant and one month, plus the master data: capacity, inverters, sensors, and anything that changed in the month.
Do we have to change our monitoring?
No. Nothing changes in your monitoring and there is no migration. The portals keep doing what they are good at.
What do we get back?
Performance ratio and availability as a base row with the rules written down, the difference against each reported number with what explains it and what stays open, a method card next to every number and a 30-minute walkthrough.
Can you always explain a portal's number?
No. Portals rarely publish their rules and may set them per plant. We test which choices reproduce the number, say how sure we are, and show what stays unexplained.
Do you recommend a portal or an operator?
No. We check the numbers. We recommend no portal, no operator and no strategy, and we never take commission from a vendor.
What if our portal does not offer an export for everything?
Tell us which portals you use in the form. That is why the field is there, and it is where the conversation starts.