Backframe · a new practice · forming 2026

The world is after your profits.

ExpectationsProcessResults

You are paying for software you don’t use, and paying again in the time it takes to work around it. The admin — the part of this business that ought to be the easiest thing in it — has quietly become the most expensive part of it. We build the process around the leading platforms in construction, and we build it to protect your margin.

Target client · annual volume · log scale
$3M $10M $50M CASE BY CASE BEST FIT
Five things that move · in this order
↑ BuildMarginBecause you can finally see it per job, per division, while the job is still running.
↑ GrowProfitThe same volume, priced and coded against what things actually cost.
↑ GrowCashBill faster because the numbers are ready, not because you chased them.
↓ ReduceOverheadSeats nobody opens, systems nobody retired, work nobody should be doing twice.
↓ ReduceHeadachesThe month-end that took two weeks. The report nobody trusts. The argument about whose number is right.
02 The bleed

Enough is enough

Before you forward this

If your instinct is to send this to whoever “handles the software,” that instinct is the problem we’re describing. Nobody delegates the decision to bid a job at cost. This is the same decision, spread over five years.

Your people aren’t the expense. The gap they’re paid to stand in is. That gap got there because at some point it looked like an admin matter, and admin matters get handed down.

Nobody set out to build it this way. It accreted — a seat here, a hire there, a system nobody retired — and every piece of it was a reasonable decision on the day it was made. Added up, it is now somebody else’s revenue, taken out of your margin, and there is no line on your P&L that says so.

SeatsYou are paying per-user on four platforms, sized for the org chart you had two years ago. Nobody audits it, because auditing it isn’t anybody’s job.
The workaroundThe most expensive integration in your company is a person retyping a number from one screen into another. Not because they’re slow — because nothing connects. It never shows up as a line item, it never gets faster, and it quietly caps how much work you can take on.
OverlapYou bought the new system and kept the old one running “just until.” That was three years ago and you are still paying both.
CloseIf month-end is a two-week project, you are paying overtime for the privilege of finding out what happened six weeks ago — too late to do anything about the job it happened on.
AdviceYou pay for the same advice twice: once to whoever scoped it, and again to the staff who quietly work around what got built.
Tool C-01  ·  What the gap costs Your numbers, not ours
$
%
hrs
$/hr
What the gap costs you a year
$54,240
Unused licences $16,800 The workaround $37,440
LineAnnualShare
Seats paid for, not opened$16,80031%
Hand-moving data between systems$37,44069%
The gap, annually$54,240100%
$271,400
The same gap over five years, if nothing changes. This sits on top of payroll — it is not your people’s wages, it is what the space between your systems bills you for standing there.

Every figure above is arithmetic on what you typed, and none of it is payroll. We are not going to promise you a percentage back — we will put this number in the engagement and you will watch it move.

03 Fit

Who this is built for

Residential general contractors, custom home builders and high-end remodelers doing between three and fifty million a year. The work fits best in the ten-to-fifty band, where you have enough volume that bad job-cost data is expensive and enough staff that a spreadsheet workaround becomes eight people's habit.

You’ll get value here

  • You bill cost-plus, GMP, or a mix, and margin depends on coding costs correctly
  • You already own good software and use maybe a third of it
  • Two systems hold the same number and the numbers disagree
  • Month-end job cost lands weeks after the month does
  • You’re migrating off desktop accounting or a legacy system like Sage
  • Good people are spending their week being the integration between two systems

You’re better off elsewhere

  • Under three million — the platforms’ own onboarding is enough
  • Heavy civil, industrial, or public works with certified payroll at scale
  • Shopping for the cheapest license reseller
  • Looking for someone to pick software and leave before anyone uses it
  • Not willing to change how the field and the office actually work
04 Failure modes

Why the implementation you already paid for didn’t take

Seam — nobody is accountable for the hand-off

Each vendor's onboarding team is measured on their own product going live. JobTread will build your cost catalog. Intuit will map your chart of accounts. Adaptive will route your invoices. Four correct configurations that don't add up to one company, because no single person was responsible for what crosses between them.

Decision — cost codes are strategy, not setup

How you break a job down determines everything you can ever learn from it: which divisions actually make money, which estimator is optimistic, which sub is worth the premium. That is an owner's decision. A setup specialist can't make it for you, and a bulk import will cheerfully lock in the wrong structure for the next five years of history.

Adoption — configured in a month, adopted in a year

The technical build is the small half. The real work is translating the system into what a project manager does on a Tuesday, what the bookkeeper does on the 5th, and what a lead carpenter does from a truck in the rain. Nobody in your office has a spare year for that, which is why the license auto-renews and the spreadsheets never die.

05 The seams

The arrows are the product

Here is where it actually leaks. JobTread implements JobTread. Intuit implements Intuit. Adaptive implements Adaptive. Every vendor stops at the edge of its own product, and the edges are where the money goes.

Here is the stack we run, and the stack most builders in this range end up with. The boxes are the easy part — each vendor sells and configures its own. What we build and own is everything drawn in orange.

HighLevel leads · homeowners subs · suppliers JobTread estimates · budgets schedule · job cost Intuit Enterprise GL · WIP · payroll multi-entity Adaptive AP · approvals docs · lien waivers customer + vendor records, once cost-coded actuals committed cost vs. budget approved bills + payments FIG. 1 — TYPICAL STACK, $10M–$50M RESIDENTIAL GC
Read the arrows, not the boxes. A vendor demo shows you a box. Your margin lives in whether a sub set up in HighLevel arrives in JobTread as a vendor and in Intuit as a payee — with the same name, the same terms and the same cost code — without anyone retyping it.
Deliverable D-01  ·  issued in week three
The Order of Precedence

Every construction contract you have ever signed contains this clause. It exists because the drawings, the specifications and the agreement will eventually disagree, and arguing it out on the day costs more than deciding it in advance. So the contract states, up front, which document governs.

Your software has the same problem and no such clause. Four systems, each perfectly capable of holding a vendor, a cost code, a budget or a payment — and no written answer to which one is right when they differ. So we write one. One page, decided by you, that every future argument gets settled against.

  • A new sub is added…HighLevel is where they are born. Everywhere else receives.
  • A cost code is created…JobTread governs. Intuit mirrors, never originates.
  • Committed cost disagrees…JobTread governs until the bill is approved, then Intuit.
  • A vendor’s terms differ…Intuit governs. It is the one that pays them.
  • Two numbers, no rule…That is a gap. It goes on the list and gets decided.

It is the shortest document we produce and the one clients pin up. In a business where four systems overlap, the thing that protects your margin isn’t another system — it is a written answer to which one wins.

06 Scope of work

Three ways in

The three engagements are the framework, in order. You can start at any of them, but you can’t skip one and get the next.

SOW / A · Expectations2–3 weeks
fixed fee

Stack Assessment

We map what you have, what it costs you, and the order to fix it in — and we hand you the Order of Precedence. Yours to keep whether or not you hire us for the build.

  • The Order of Precedence — which system governs which decision
  • Current-state map of every system and hand-off
  • Cost code + chart of accounts architecture
  • Where hours and margin leak, quantified
  • Sequenced 90-day plan with owners
SOW / B · Process60–120 days
fixed scope

Implementation

We build it and get it used. One person accountable for the whole stack, including every seam between platforms.

  • Cost code architecture + budget templates
  • QuickBooks or Intuit Enterprise Suite alignment
  • Adaptive AP and approval routing
  • HighLevel as the sales and vendor front door
  • Legacy migration, including Sage job-cost history
  • Role-by-role SOPs and training, office and field
SOW / C · Resultsmonthly
retainer

Fractional Systems Ops

For companies that need the systems owned, not just installed. We stay on as the person whose job this is.

  • Month-end job cost close and WIP
  • New-hire onboarding into the stack
  • Workflow changes as the business changes
  • Vendor, integration and release management

Pricing is scoped per engagement and quoted before work starts. We hold partner status with the platforms we implement and earn on licensing as well as on our work — see question three.

07 In fabrication

What we’re building to sell, not just to bill

Every tool here exists so that a job gets done without a person being hired to do it by hand. Consulting that only sells hours stops the day you stop. Every implementation we run leaves behind something reusable — a connector, a template set, a migration tool — and those become products in their own right. Two of these already run inside our own company, which is why they exist at all.

ItemPlatformStatus
Sage → JobTread migration tooling — crosswalk-driven cost-code remapping, so legacy job history survives the move instead of being abandoned Sage → JobTread Running at MDB
Builder’s HighLevel snapshot — pipelines, custom fields, intake forms and workflows preconfigured for a residential GC, deployable in an afternoon HighLevel Running at MDB
Cost-coded actuals connector — committed cost, approved bills and payments moving between the field system and the ledger without a person in the middle JobTread ↔ Intuit In build
Residential cost-code library and budget templates, structured by project type — new build, gut renovation, addition, service work JobTread In build
Month-end close pack — the WIP schedule, the job-cost review and the checklist that makes closing a two-day job instead of a three-week one All Spec’d

Where a platform runs a marketplace, that is where these will live. A tool you can buy without buying us is the point.

08 Submittal log

What we did to ourselves first

Every item below is our own company, not a case study we bought. The statuses are current and honest — some of this is still in flight, which is the point: you are hiring someone doing the work now, not someone who did it once in 2019.

ItemSystemStatus
Legacy Sage job-cost history migrated into JobTread through a purpose-built cost-code crosswalk Sage → JobTread Complete
AP and document management moved off hh2 onto Adaptive Adaptive Complete
Three-axis cost and billing model for cost-plus and soft-GMP contracts JobTread In production
Chart of accounts and job costing aligned to Intuit Enterprise Suite Intuit Enterprise In progress
HighLevel as the front door — leads, customers and vendors syncing into JobTread as one record HighLevel → JobTread In progress
AI voice agent answering the main line and routing after hours Telephony In production
Written SOPs per role across the whole stack, office and field All Ongoing
09 Questions

The five you’re about to ask

Do you make money on the software?

Yes, and you should know exactly how. We hold partner status with the platforms we implement, so we earn on licensing alongside our fees. What that buys you is real: direct escalation into the vendors when something breaks, and visibility into what’s shipping before it ships. What it costs you is a conflict worth testing — so on the first call, ask us what to keep and what to drop. “Keep what you already own and configure it properly” is an answer we give constantly, and it pays us nothing.

Can I hand this to my office manager?

No — and we’ll decline the engagement if that’s the arrangement. Not because your office manager isn’t good at their job; because the decisions in the first three weeks are yours. How a job breaks down into cost codes determines what you can ever learn about your own margin, and no employee can or should make that call for an owner. Bring them in on week four and they’ll be your best asset on the project. Start with them and we’ll build a very tidy version of the thing you already have.

You take commissions. Doesn’t that make you part of the problem?

Press on it. We earn on licensing, so we have an obvious incentive to sell you more seats — which is exactly why you should measure us on the opposite number: what your combined software and admin spend looks like a year after we finish. We’ll write that number into the engagement and put it in front of you at close. Partner status exists so we can escalate into the vendors on your behalf and see what’s shipping before it ships. It is not a reason to keep you paying for seats nobody opens.

We already bought the software.

Good — that’s most of our work. The common situation isn’t a bad buying decision, it’s a correct purchase that was configured to a demo instead of to your company.

You’re a builder. Do you compete with us?

We build on Cape Cod and the Islands. If you work in our market we’ll tell you on the first call and you can decide. Outside it, we’re just a contractor who happens to understand your P&L.

Who actually does the work?

The people who did it at Duffany Builders. You won’t get handed to a junior implementer after the sales call, because there isn’t one.

What if we’re on Buildertrend, Procore or CoConstruct?

Fine. The seam work is the same problem in different clothing, and we’ll say plainly whether the platform is worth keeping. We go deepest on JobTread, Intuit Enterprise Suite, QuickBooks, Adaptive and HighLevel.

10 Intake

Get margin back.

Tell us what’s broken. Eight questions, about two minutes, and we’d rather the owner filled it in. We read every one. If it’s a fit you’ll hear back with a short read on your situation and a link to book a call — not a sequence of drip emails.

Form F-101  ·  Readiness intake Rev A  ·  Sheet 1 of 1

No newsletter, no drip. One reply from a person.

Practice
Backframe · forming
Framework
Expectations · Process · Results
Market
US residential GC · $3M–$50M

Backframe · noun  The blocking a carpenter sets behind drywall before it closes up, so that later you can hang a cabinet run, a grab bar or anything heavy exactly where you need it. Nobody ever sees it. Everything you want to hang afterwards depends entirely on whether somebody put it there.