Fictional client
Wattlebank assigns the certificates on every job and discounts their value off the customer’s price on the day they sign. The certificates are created after the install, traded weeks later, and can be knocked back months after that — for a missing photograph, an unsigned declaration, a serial number nobody could read. Seventy-four jobs were rejected or withheld last year, worth $211,000. The customer had already had the discount. The panels were already on the roof. Solar is a business where the software either captures the evidence at the moment of install, or the money goes back.
Wattlebank Energy does not exist. We cannot show live client systems, so this study is built around a fictional installer — invented business, invented crews, invented customers, invented figures. The network operator, the retailers and the certificate prices are all invented too. What is not invented is the work: the evidence chain, the accreditation obligations, the grid connection and meter mechanics and the failure modes are all real, and the screens are genuine working pages rather than pictures of pages. Every mockup on this page opens and responds to clicks.
On scope: this is written for residential solar and battery installation in Australia, with a South East Queensland firm as the example. Certificate values, deeming, state incentive schemes and network connection rules all change, and several differ by state — in a real build those are configuration rather than code, for exactly that reason. All figures are Australian dollars.
Wattlebank fits residential solar and batteries across Brisbane, Ipswich and Toowoomba. Five crews, four accredited installers, 1,180 systems a year and a genuinely good reputation. It is also running at 22.8% margin in a market where everybody is, which means the difference between a good year and a bad one is not sales — it is how much of what it has already earned it manages to keep.
None of these is a sales problem. Every one of them is money Wattlebank has already earned and then hands back, usually because of something that would have taken thirty seconds on the day.
One rule holds this together, and it is deliberately blunt: a crew cannot close a job on the tablet until the evidence pack is complete. Not a reminder, not a report somebody runs on Friday — the same hard stop as testing the system before you leave. Serials are scanned rather than photographed, so an unreadable one is rejected on the spot. The array shot prompts from height with an example on screen. Check-in and check-out are geotagged, which takes four seconds and removes the attendance question entirely.
Everything else follows from that. Certificates get created in days rather than weeks, which removes most of the price exposure without anybody taking a view on the market. Return visits stop, which gives back fifty-one installs a year. And the audit sample stops being something to dread, because the evidence was always going to be there.
Eight screens follow. Each one opens as a real page — click through them.
An enquiry that landed at 21:47 on a Tuesday, answered in thirty-eight seconds, qualified over four exchanges, and holding a Saturday assessment slot by 21:54. Bill, daytime occupancy, roof type, ownership, postcode and certificate zone are all captured on the way through, so the assessor arrives already knowing them.
Underneath, conversion measured against how long the customer waited, and the channel breakdown — including the one enquiry in seven that is a missed call, better than a fifth of which were never returned.
Where the certificate value actually sits — traded, created, ready to create, evidence incomplete, or gone. $112,000 is in jobs where the pack is complete and nobody has created the certificates yet, which is a queue with nobody assigned rather than a compliance problem.
Alongside it, grid approvals by connection type and the nine jobs booked to install before their approval is back.
The screen this study is built around. The customer got $2,850 off the price on the day they signed. The certificates are created after the install and traded weeks after that — and this job cannot create them at all, because six panel serials are unreadable, the battery serial was never captured, and the array photograph was taken from the driveway.
None of it can be fixed from the office. A job that was worth $2,180 of margin at quote becomes a $1,010 loss on a photograph, and fifty of the seventy-four failures last year were something a phone could have captured correctly on the day for nothing.
Wattlebank discounts at one price and sells at another, six to ten weeks apart, and nothing connects the two. Over twelve months that gap cost $47,000 — not a trading position anybody took, just a number nobody watched.
Then the finding that reframes the whole business: $170 a job goes on return visits for evidence. Across 1,180 installs that is $200,000, almost exactly the $211,000 lost to rejections. The same failure charges twice.
One installation end to end — enquiry, design, grid application, the approved export limit, install, certificates, meter. Sixteen of the eighteen documents on this job are perfect. The two that are not are the two that decide whether $2,850 exists.
It also tracks what is still owed to the customer: forty days after install, this household is exporting to the grid and not being paid for it, because the retailer has not reconfigured the meter.
Five crews, and four of this week’s twenty-three slots are return visits to collect evidence — most of a crew-day, every week, invisible in the accounts because it is not booked as anything.
Over a year that is 184 return visits, $62,400 of direct cost, and fifty-one installs that did not happen. The displaced margin is $109,000, which is the number that actually matters.
Everything Wattlebank sells rests on an accredited installer physically on the roof — not supervising from the van, not signing off afterwards. Seventy-four per cent of jobs evidence that with a geotagged check-in. Nine per cent have a signed job sheet, which proves a signature and not attendance.
All four jobs with nothing on file are the subcontract crew, who are paid per install and whose paperwork nobody inspects on the day.
A second audience. The job in plain terms, the documents worth keeping, and the two things a solar customer actually rings about: why the export limit is lower than they were quoted, and why they are not being paid for what they export.
It says both plainly, including that the meter is the retailer’s job rather than Wattlebank’s — and then offers to chase it anyway.
Four phases of two weeks each — eight weeks end to end. The on-site capture goes live first, because every install that happens without it is another job that may come back. Everything behind it is reporting on data the crews are already producing. Each phase ends with something the team actually uses, not a demo.
Weeks 1–2.
Weeks 3–4.
Weeks 5–6.
Weeks 7–8.
Eight weeks is possible because most of it already exists. The job and customer records, document handling, obligation and expiry tracking, escalation rules, the messaging engine and the portal are running in production across six sectors already, including installer businesses that live or die on evidence captured in the field. The conversational SMS and email layer behind the first-response screen is a product we already sell. What actually gets built for an Australian solar firm is the sector core: the evidence requirements per system type, the certificate lifecycle, and the network connection gating. That is the part this study is about, and it is the part we would spend the eight weeks getting right.
This took a fictional installer and made its process visible. The same exercise works better with a real one. Tell us how your business actually runs and we will build a working mockup of it — your systems, your crews, your brand — before anybody commits to a build.
Start a Conversation See other sector blueprints