About

Nouman Sadiq — business systems architect for general contractors.

What gets built here

Back-office software for general contractors, residential and light commercial, and the trades that work alongside them. One system for the office, the crew and the owner: quotes priced within the limits you set and followed up on the days you pick, one page per job, extra work priced and signed before the crew starts, receipts and photos from the field on the right job, what every job left you after subs and materials, invoices and who owes you, and sub payouts ready for your OK.

The rule behind all of it: your company, run on your rules. You make the calls; the system carries them out. It won’t make decisions for you — it only carries out yours.

The companies this suits have a particular shape. The work is real and the demand is there. The constraint is that too much of the operation is held in one person’s head, or re-keyed between three tools, or tracked by memory against a deadline. That is an operations problem and it has a build-shaped answer.

If the real constraint is demand rather than operations, a system will not fix it. That is worth saying on the first call rather than after a deposit.

Checkable, not claimed

The live demo. A sample general contractor you can click through end to end, from the quote to the sub payout, on clearly marked sample figures: open the live demo.

The Texas lien deadlines. A free calculator that returns every notice, lien affidavit and lawsuit deadline under Texas Property Code chapter 53, each one citing and quoting its statute section, with the rules published as an open dataset: Texas lien deadlines.

What each system does on a bad day. The failure modes are written down, with what each one deliberately does not do: how it’s built.

The method

Study the business. Build the thing that is breaking. Run it on sample data until it is real.

Study the business first. Not the software category — the business. Where work queues. What gets re-keyed. Which step everyone has quietly built a spreadsheet around. Which deadline is tracked in someone’s head. That hour decides whether anything after it is worth doing, and sometimes it decides that it is not.

Then build the specific thing that is breaking. Not a platform, not a suite. The one process that costs the most and fails the most often, built properly, so it stops producing exceptions. A second system comes after the first one is running and has earned its place. Systems that try to be everything on day one tend to be finished by nobody.

Then run it on sample data until it is real. Every screen is populated with invented but realistic records and clicked through before it goes near a live one. That is how the shape of a thing gets found: the field that is missing, the state nobody planned for, the screen that reads well in a diagram and badly in use.

The part of this that is hard to fake is the failure-mode table — what a system does on a bad day and the honest cost of each answer. Those are published, per system, at /architecture, alongside what each one deliberately does not do.

What a project looks like from your side

A free 30-minute call first, and one fixed price out of it. The whole office is $6,500, or $500 a part on a $1,500 setup, and larger multi-part systems are quoted as one fixed price. The price is set against what the system has to do, not against a tier or a seat count.

A 25% deposit reserves the build slot. It is fully refundable if, after the requirements call, either of us decides this is not the right fit — and if we go ahead, it is credited in full toward the project. Nothing is built from a conversation alone: the deposit pays for the real mockup of your operation, your written approval of it becomes the scope, and the scope is what the delivered system is tested against.

A core build takes roughly three weeks from an approved scope; a larger system has its timeline written into the scope. Handover includes testing against the written scope and 30 days of defect support at no additional cost.

The terms behind all of that are published rather than described: Terms of Service, Refund Policy and Privacy Notice.

Custom builds, not resold software

Nothing here is somebody else’s licence resold with a markup. There is no per-seat fee, no tier that moves when you hire someone, and no vendor sitting between you and your own records.

Each build runs in its own container against its own database with its own credentials, rather than sharing tables behind a filter. Integration credentials are registered under your account and authorised by you through the provider’s own flow, so you can revoke access without asking anyone. On final payment the delivered system is yours to use in your business for good, and your data is yours to export at any time. It runs on infrastructure I maintain, which is what the monthly Care subscription pays for. Reusable framework components stay part of the practice and get reused across projects, the way any studio keeps its toolkit.

The trade is worth stating plainly. A custom build costs more than renting a product, and if your process already fits one, rent it. It is worth paying for when it does not — when your payout rules, approval chain or a crew that works in another language live in a spreadsheet beside the software. It takes weeks to stand up rather than an afternoon, and in exchange the system fits the business instead of the business fitting around it.

Every system on this site runs on sample data

Every figure in every console on this site is invented. The bookings, the queues, the names, the revenue lines — all of it is sample data, shown to demonstrate the interface. None of it is real client data and none of it is a performance claim.

That is a principle rather than a disclaimer. Somebody else’s real numbers prove nothing about what your system would do, and publishing a client’s live operating data in order to sell to their competitors is not a trade worth making. So the systems are built out fully, populated with realistic records, and left clickable. You use the interface instead of looking at a picture of it.

What is real is the software. The flows work, the states are handled, and the failure modes are written down. The rule that follows from it is simple: nothing invented is ever presented as real, and where a number would be a claim, there is no number.

Questions people ask first

What does a system cost?

It starts with a free 30-minute call: you tell me how the work actually moves, I show you the closest system already built, and you get one fixed price. The whole office — all ten parts — is $6,500 to set up and $450 a month. Or start with only the parts you need: $500 each on a $1,500 setup, Care from $20 a part. You pay in four 25% stages, each before it starts. That deposit pays for a clickable version of your own system, built from your requirements, and your written approval of it becomes the scope. The twelve extras on the shelf — client portal, crew app, QuickBooks sync and the rest — are $500 each plus $20 a month; a new extra built for your company alone is priced at its size, $1,500, $2,500 or $4,000. Bigger companies — several locations, entities or departments — are quoted as one fixed price. After that, a monthly plan keeps it running, priced by what it includes rather than how many people use it: Care is $450 a month (hosting, backups, security, bug fixes, next-business-day replies and up to 2 hours of changes), Grow is $850 (same-business-day replies, up to 6 hours of changes and a quarterly review) and Partner starts at $1,500 (up to 12 hours, a monthly planning call and 2-business-hour replies on urgent issues). Paying a year up front gets two months free.

How long does a build take?

A core build takes roughly three weeks from an approved written scope. Larger systems take longer, and when they do, the timeline is written into the scope before work starts rather than discovered halfway through.

What happens to my data?

Each build runs in its own container against its own database with its own credentials, rather than sharing tables with anyone else. Integration credentials for Google, a payment processor or a messaging service are registered under your own account and authorised by you through the provider's own flow, so access can be revoked at any moment without asking us. Business information shared during a build is treated as confidential. A full data export is available on request at no cost, including if you leave.

What if I already have software?

That is the usual starting position, and most of it stays. A system is built to sit alongside the tools that already work and replace only the step that is breaking — the accounting ledger stays the ledger, and a tool with a usable API gets integrated rather than replaced. Sometimes the study finds that the software already owned covers the problem and the fix is configuration rather than a build. That answer is given plainly when it is the true one.

Who owns the system once it is built?

You do. On final payment the delivered system is yours to use in your business for good, and your data is yours — exportable at any time at no cost, including if you leave. It runs on infrastructure I maintain, which is what the monthly Care subscription pays for, so you never look after a server. Reusable framework components stay part of the practice and get reused across projects, the way any studio keeps its own toolkit.

Making contact

Bring the thing that goes wrong on a bad week.

The job re-keyed into three places, the report that takes a day to assemble, the deadline nobody owns. That is the useful place to start a first conversation, and it is more use than a list of features.

Book a 30-minute call