SPARKYQC
← Back to Sparky QC
Last updated 28 July 2026
Written for the people who have to stand behind the package. This is what the system actually does with your records — including what it checks, what it does not check, and where the judgement stays with you.
Sparky QC has 41 record sections — cable schedule, cable pulls, terminations, conduit and raceway, cable tray, grounding, panels and circuit schedules, transformers, motors, lighting, heat trace, insulation resistance, continuity, ground resistance, loop checks, functional tests, relay testing, torque records, phase rotation, calibrations, qualifications, permits, punch list, NCRs, RFIs, observations, and the rest.
They are not 41 separate systems. Every section stores its rows in the same underlying structure: a record belongs to a project, is tagged with which section it is, and carries its own fields. That is why every section behaves the same way — the same table, the same search, the same CSV export, the same print view, the same attachment handling — and why a new field can be added to a section without disturbing anything already entered.
It also means your data is never trapped in a section. Everything exports, and Save Snapshot on the Project Turnover tab downloads a complete copy of a project that is yours outright.
Records are linked by the identifiers you already use on paper — cable number, equipment tag, circuit reference, loop number, instrument ID. You type them once, and the same value is what ties the cable schedule to the termination record to the megger result to the calibration certificate behind it.
The clearest example is a cable. The cable number on the schedule is the same value on its pull record, on both termination records, and on the megger result — so the insulation test carries the cable's size, voltage class and routing without any of it being retyped. Neither tab has to hold the other's data, and neither goes stale when the other is edited.
Consistency matters more than format. The links are only as good as the identifiers. If a joint is W-104 on the map and W104 in the log, they are two different joints as far as any software is concerned — including this one. Agree a format at the start of the job and hold to it.
The audit runs three families of check across every section. It is deliberately narrow: it reports things that are objectively true from the records in front of it, and does not guess.
| Check | What it looks at | Result |
|---|---|---|
| Expiry | Any section with an expiry date — electrician qualifications, NFPA 70E cards, equipment certs, instrument calibrations, permits, one-calls | Blocking if the date has passed, Review if it falls within 30 days |
| Still open | NCRs not Closed; RFIs not Closed; punch list items not Signed Off |
Blocking for an open NCR, Review for the other two |
| Missing document | 13 sections whose whole purpose is holding a document — qualifications, calibration certificates, permits, procedures, insulation and ground resistance reports, relay and functional test reports, subcontractors and others | Review when every document slot on the row is empty |
Each finding names the section, the record, and the reason, and clicking it opens the record. The list exports to CSV and prints as a report on the same letter shell as the job book.
Being clear about this is more useful than overselling it:
An empty audit is not a statement that your project is compliant. It means nothing in the records is expired, unresolved, or missing its attachment. Whether the work meets code is determined by your procedures, your qualified people, and the governing specification.
Alongside the findings, the audit shows a record count for every section. A zero is not flagged as a problem — most jobs do not use every section, and a system that nagged about empty tabs would train you to ignore it.
The value is in reading it yourself: a section you expected to be populated and which shows zero is worth asking about before handover, not after.
The job book assembles your records into a single printable document in turnover order — the same order a client expects to leaf through. It includes a binder cover insert, section dividers, spine labels, and an index at the end.
It is generated from the records at the moment you print it. There is no separate "compile" step to forget, and no chance of the book disagreeing with the tabs it came from.
Where a section supports it, Read from document extracts values from a certificate or report and fills in the form for you.
Details of who processes what are in the Privacy Policy.
Access is enforced by the database itself, not by hiding buttons in the interface. A user with no relationship to a project cannot read or write to it, and that holds whether they come through the app or anywhere else.
Nothing is deleted. Your account becomes read-only for 90 days: sign in, read everything, print job books, export CSVs, download attachments. A complete export is emailed to you automatically when the subscription ends, with a link valid for the whole period.
Resubscribe inside those 90 days and everything is exactly where you left it. Full terms are in the Refund and Cancellation Policy.
Sparky QC organises, tracks, and compiles records, and tells you when something in them is expired, unresolved, or undocumented. It is a filing system that pays attention. That is as true of a calibration certificate as it is of a megger result.
It does not certify, approve, or verify anything. A QA/QC record is a legal document, and the signature on it is yours.
Questions: support@sparkyqc.com. See also the Terms of Use.
How Sparky QC works — sparkyqc.com/how-it-works