Ask a skylight cleaning robot supplier seven things before signing in 2026: real climb angle on wet glass, grip at panel joints, consumable life in hours, spare-parts lead time, on-site training, service response time and a reference site you can call. Vague answers to any of them are a signal.
Most bad purchases were visible in the sales conversation. The supplier dodged a question, the buyer accepted a rounded number, and the problem surfaced six months later on a wet roof. These questions are designed to make that harder.
Why do these questions matter so much?
A skylight cleaning robot is a five-year commitment, not a one-off tool. Choices made at the quote stage — grip, filtration, service — decide your running cost long after the salesperson has moved on. Asking hard questions up front costs nothing and saves entire seasons of frustration.
What should the supplier prove on grip and pitch?
Ask for the tested climb angle on wet glass, not the marketing figure. Many units claim steeper pitches than they hold once the surface is slick. Then ask specifically about edge grip: what happens at a panel joint, a raised frame or a mullion where the surface changes? A supplier who knows the machine can describe the failure point honestly.
- Tested angle dry and wet, quoted separately.
- Behaviour at panel joints, frames and transitions.
- Minimum glass thickness the machine is rated for.
How do you test the water and spotting claim?
Streak-free glass depends on filtration, and filtration depends on site water. Ask for the filter’s expected life on your specific water hardness, and for a spot-free guarantee in writing. If the supplier cannot state your water hardness limit, run your own test: clean a sample panel and let it dry for 24 hours before judging.
What service commitments are non-negotiable?
| Commitment | Reasonable 2026 standard |
|---|---|
| Spare-part lead time | Under 10 working days |
| On-site operator training | Two people, at handover |
| Fault response | Defined hours, not “soon” |
| Consumable life guarantee | In hours or m², in writing |
A ten-day part lead time sounds fine until it lands in the middle of your only clean window. Get it in the contract.
Which red flags end the conversation?
- No reference site you are allowed to call or visit.
- Consumable life given only as “it depends” with no range.
- No mention of CE documentation or IP rating.
- Pressure to sign before any roof trial.
- Training treated as an optional extra.
One caveat in the other direction: a supplier who says your roof is a poor fit for a robot is usually being straight with you. A machine sold into the wrong site is a refund, a support headache and a bad reference for them.
Bring these seven questions to your next call. Machine specs, documentation and site-fit advice are available through the contact page.
What should the warranty actually cover?
Ask whether the warranty covers water ingress, motor wear and battery capacity loss, or only manufacturing defects. On a wet glass roof the first two are the likely faults, and a defects-only warranty leaves you paying for them. Get the coverage in writing, including the labour rate if the machine must be sent away.
How do you check the consumable cost claim?
Ask for the price of one blade, one brush set and one filter, then multiply by the life figures they gave you. That gives an annual consumable cost you can compare across suppliers. A machine with a low purchase price and expensive, short-lived parts is often the worst value on the shortlist.
What question exposes a weak supplier fastest?
“What is the failure mode you see most often, and how do you fix it?” A supplier who knows the machine answers instantly with a specific part or condition. A supplier who sells boxes deflects to features. That single answer tells you more than an hour of product talk.
How do you compare two suppliers fairly?
Build a one-page table: purchase price, tested wet climb angle, tank size, runtime, blade life, part lead time and training included. Fill it from written answers, not calls. The supplier that scores well on the boring rows is usually the one still answering the phone in year three. Demo videos do not belong in the table.
What if no supplier fits your roof?
That is a real outcome, and better found now. Some roofs — very steep, heavily segmented or structurally weak — are not good robot candidates yet. In that case a rope crew plus a revised schedule is the sensible answer. A good supplier will tell you that rather than sell you a machine that keeps stalling.
What should be in the final contract?
Purchase price, delivery date, the tested wet climb angle, consumable life figures, spare-part lead time, training for two operators and a named service contact. Put numbers on each row. A contract that reads like a spec sheet is easier to enforce than one full of adjectives, and it protects both sides when a fault appears mid-season.
What is the one question people forget?
“How do I get it off the roof if it fails?” Recovery is the question almost nobody asks and everybody needs an answer to eventually. A stalled machine on a glass roof is a logistics problem, and the answer should be in the handover pack before the first fault, not improvised when it happens.
Key Takeaways
- Demand climb angle on wet glass and behaviour at panel joints, not marketing figures.
- Tie filter life and spotting claims to your actual water hardness.
- Get spare-part lead time under 10 working days in the contract.
- A supplier who declines a bad-fit roof is a good sign, not a lost sale.