A 90-day rotation works on average — but no farm operates in averages. Some sites need attention sooner; others are still clean when crews arrive. The schedule is necessary; the schedule is also imprecise.
SCADA detects underperformance after losses materialise — that's its job. Soiling events typically build for 1–3 weeks before the production curve drops far enough to flag. The lag is structural, not a monitoring failure.
Across 20 sites with one cleaning crew, weekly urgency varies enormously — but most O&M teams lack a consistent objective ranking across farms. Decisions default to schedule, geography, or experience.
Dispatcher pulls up the cleaning schedule. Crews go where the calendar says.
"Production gap looks soiling-related. Last cleaned 11 weeks ago, next on the schedule in 2 weeks." Reasonable answer — but no objective measurement to go with it.
Cleaning budget spent at plan, soiling losses present in production data, no quantified link between the two. Defensible to ownership only on the basis of "we ran the schedule."
Dispatcher opens the priority list. Sites 7 and 12 ranked highest this week — both projecting >3% production loss. Crews go there first.
"Site 7 was cleaned Tuesday — production back to baseline. We caught the soiling event 11 days before SCADA would have flagged it."
Fewer cleaning runs, lower soiling losses, defensible audit trail of every cleaning decision. Cleaning budget down 20–40%.
200,000 MWh/yr at $40/MWh. Soiling loss data: Wood Mackenzie 2025. SkySense: $18K/yr (Professional tier).
| Cost line | Without SkySense | With SkySense |
|---|---|---|
| Manual inspections / year | $6,000 | $0 |
| Cleaning cycles / year | 4 × $5K $20,000 | 3 × $5K $15,000 |
| Soiling production loss | 5% $400,000 | 2% $160,000 |
| SkySense subscription | — | $18,000 |
| Total annual cost | $426,000 | $193,000 |
Across a 500 MW portfolio under O&M management → ~$1.1M in addressable annual savings, with a SkySense subscription cost of ~$72K/year.
Net annual savings, 100 MW farm
Every site you manage, ranked by soiling urgency and revenue impact this week. Direct feed for your dispatch team.
For each flagged site: estimated production loss per week, cleaning crew cost, projected payback. Defensible to a CFO or ownership.
Branded report your account team can hand to the asset owner — soiling status, cleaning actions taken, production loss attribution, trend vs. prior month.
SkySense is a layer above your existing stack, not a replacement for it. Your SCADA continues to do what it does — collect production data and surface performance metrics. SkySense adds the soiling-specific intelligence your SCADA platform doesn't have: forward-looking detection, ROI scoring, and portfolio prioritisation. We integrate with kWh Analytics, AlsoEnergy, Locus Energy, and most major SCADA platforms via API or webhook — soiling alerts can flow into your existing dispatch workflow rather than living in a separate dashboard. CMMS integration (ServiceMax, Maximo, etc.) is available on Business and Enterprise tiers.
Not from a SkySense perspective. The output of our system is a prioritised list of which farms need cleaning attention this week — what happens operationally after that is your team's call. In practice, most O&M companies we talk to use SkySense to make smarter dispatch decisions to their cleaning subcontractors: fewer trips, better-targeted, with documented evidence of why each cleaning was warranted. Some clients use the priority list and ROI scores directly inside their cleaning contractor SOWs to make the work scope explicit.
Different question than people often realise. A site walk is a snapshot of one moment, by one person, looking at a sample of panels. It's accurate for what it sees but limited in coverage — you can't walk every row of a 100 MW farm every week, and the inspector probably isn't at every site this week anyway. SkySense covers every panel row across every site you manage, every 5–7 days, with the same algorithm — meaning the comparison isn't accuracy-per-observation, it's information-per-month across your portfolio. During the pilot, we encourage clients to run on-site validation on at least one or two flagged sites to directly compare our soiling estimates against observed conditions. That builds the trust faster than any spec sheet from us would.
More questions?
Talk to our team before you apply →Structured 90-day pilots with a small number of O&M companies at a time. Your farms, your SCADA history, your cleaning schedule — analysed against actual satellite imagery to produce results you can take to ownership.
Apply for a Pilot ProgramWe'll confirm fit and walk you through what the 90 days involves before anything starts.