Custom integration

Sold is not the sameas installed.

The proposal was agreed with the client by the owner or the sales designer. The install packet is built by somebody else. Accessories, rack layout, network requirements, trim colours, programming scope and labour assumptions all drift, and the crew builds what the packet says.

Built inside D-Tools and your project platform, not instead of them

Nothing to do with system design or programming

One system at a time, proven before we build the next

For established integration firms with several concurrent projects and a project management layer. Not a solo installer, and not retail-only.

Where the work stops moving in smart home and AV:

After the proposalThe client goes quiet while they talk to their architect, their builder, their partner and their budget. Nobody owns the next contact.
Mid projectA finish, a keypad colour, a rack location. The integrator knows it is blocking. The client does not.
At handoverThe service agreement is offered verbally if at all, and renewal depends on somebody remembering.
What we build first, and why

Four things that decideyour month.

The proposal keeps moving without you

An owner, a schedule, and a branch on the reason it stalled.

A custom integration proposal is a long, considered, high-consideration document, often revised several times. It goes out, and the client goes silent while they consult four other people.

Nobody owns the next contact, so it ages. The client eventually proceeds with whoever stayed in touch, or reduces the scope to whatever they still remembered from the document.

After the quote goes outhow it is wired

The sequence

Quote delivered, clock starts
day 0
Checking it arrived and reading right
day 2
The specific thing that usually stalls it
day 5
A decision, or a reason there is not one
day 10
Stops the second they reply

And when they do reply

“Too expensive”Goes to options, not to a discountBranch
“Not right now”Parked, and picked up on the date they gaveBranch
“Need to speak to someone”Handed to a person with the history attachedYou

Three different problems. One “just following up” message answers none of them.

Selections stop stalling the job

Deadlines set with the programme, and the client told what their decision is holding up.

A project waits on a finish, a keypad colour, a rack location, a fabric, or a decision the client does not realise is blocking anything. The integrator knows. The client does not.

Days pass, then the trade partner's schedule moves, then the project slips a month for a decision that would have taken four minutes had anybody asked for it properly.

Capture & Respondhow it is wired

Arrives on

Missed callRang out during a job9:41pm
Website formQuote request9:41pm
Instagram DMAsking about price9:42pm

Three channels, three inboxes, one person expected to watch all of them. After: one lane, whatever they used.

Then

Logged against one contact
on arrival
Reply sent from your number
+30 sec
Routed to whoever is on call
by service type
Chased until there is an answer
+1h, +1d, +3d
Booked, or handed to a person

Sold scope and install packet stop diverging

One packet from sold to installed, so the crew builds what the client bought.

The handoff from sold to installed is where expectation and delivery separate, and neither party notices until the client walks in. The crew builds what the packet says. The client expected what the proposal said.

Change orders belong here too. Somebody on site says yes to something, the work happens, and it is never documented, priced, approved or attached to a payment milestone. It is absorbed, and the owner finds it at the end of the job.

Your list, 8amcontrol

Overdue

Quote sent, no replyPast the follow-up windowChase
Consultation unconfirmedTomorrow morningConfirm

Failed, and why

Reminder could not sendNumber on file is a landlineFix

Waiting on a person

Price exception needs sign-offOutside the rule we agreedYou

The installed base becomes recurring revenue

The agreement offered deliberately at handover, and renewal that does not depend on memory.

Two-thirds of CEDIA members offer service agreements. Far fewer convert their existing installed base onto one, because the offer is made verbally at handover if at all.

Meanwhile support requests arrive by text message to the owner's personal phone, which is both the worst possible queue and a free signal about which clients would buy an agreement if asked.

The return windowhow it runs

Triggered by the work, not by a memory

Job completedThe clock starts here, automaticallyDone
Inside the windowThe point where coming back is obviousReach out
Past the windowDrifting, and worth a different messageRecover

Why it stops without this

Nobody decides to lose a returning customer. The interval passes while the team is busy, the moment to ask goes with it, and six weeks becomes ten, then twelve, then never. The window is a date on a record, so it does not depend on anyone noticing.

The build order

Seven systems forsmart home and AV.

We install one at a time, pointed at your platform and your stages, and prove it moved before starting the next. Each one has its own page: what it is, the problem it solves, how it works step by step, how it gets built, and how to measure it.

  1. Tier 1Foundational

    Multi-channel capture, qualification and instant response

    Enquiries, builder referrals and architect introductions into one queue, qualified before design time is spent.

    Why here in the order. Fastest visible result, and it captures the trade introductions that currently live in text messages, which is the finding that earns trust.

    After it goes live. Design hours go only into projects that could proceed.

    The whole system in detail

  2. Site visit preparation and consultation protection

    The visit gets scheduled, prepared and documented rather than improvised.

    Why here in the order. Protects the most expensive hour in the sales process.

    After it goes live. The first meeting on a high-trust sale stops being ad hoc.

    The whole system in detail

  3. Tier 2Growth

    Proposal follow-up and decision recovery

    An owner, a schedule that runs while you are on site, and a branch on the real reason it stalled.

    Why here in the order. Usually the single largest recoverable pool in the business.

    After it goes live. Proposals stop ageing while the client consults four other people.

    The whole system in detail

  4. Selections, change orders and milestone control

    Selection deadlines with the schedule consequence attached, and changes priced and approved before work happens.

    Why here in the order. Stops the margin loss, and is the one the project manager will champion hardest.

    After it goes live. Margin stops being absorbed on site.

    The whole system in detail

  5. Tier 3Flagship

    The complete project system

    Sold through commissioned as one connected path, with the packet matching the proposal.

    Why here in the order. The flagship, once the pieces are proven individually.

    After it goes live. The crew builds what the client bought.

    The whole system in detail

  6. Service agreements and installed base revenue

    The agreement offered deliberately at handover, and the existing installed base worked rather than forgotten.

    Why here in the order. Compounds everything above, and depends on the installed base data that System 5 starts producing properly.

    After it goes live. Support stops arriving as texts to a personal phone.

    The whole system in detail

  7. The owner's morning brief

    One list: proposals ageing, decisions overdue, changes unapproved, agreements unrenewed.

    Why here in the order. Only meaningful once there are systems to report on.

    After it goes live. The owner stops holding the whole book in their head.

    The whole system in detail

The whole map

Every stage, what breaks,and what it costs.

This is the stage map from our own research on custom integration. Two rows carry most of the money here, and the second of them is a handoff rather than a sale.

Capture

What breaks

Client enquiries, builder referrals and architect introductions land in different places.

What it costs

High-value introductions wait, or have no owner.

What should happen

One queue, with a source and an owner on every item.

Qualify

What breaks

Budget band, project stage and decision process are established late.

What it costs

Design and engineering hours on projects that were never going to proceed.

What should happen

Qualified before design time is spent.

Design

What breaks

Revisions fragment across email and nobody holds the current version.

What it costs

Work redone, and a proposal that arrives late.

What should happen

A decision log with an owner and dates.

Proposethe expensive one

What breaks

The proposal goes out and nobody owns the next contact.

What it costs

It ages, then proceeds at reduced scope or with somebody else.

What should happen

An owner, a schedule and a branch on the actual reason.

Handoffthe expensive one

What breaks

Sold scope and install packet diverge on accessories, rack, network and programming.

What it costs

The crew builds one thing and the client expected another.

What should happen

One packet, from sold to started.

Install

What breaks

Selections stall the programme and changes are absorbed on site.

What it costs

A month lost to a four-minute decision, and margin gone quietly.

What should happen

Deadlines with consequences attached, and written approval before work.

Complete

What breaks

Punch, final programming, training, documentation and final payment close unevenly.

What it costs

A job ninety-five per cent finished for six weeks with the invoice sitting.

What should happen

A completion checklist with training and collection attached.

Support and renew

What breaks

Agreements are offered verbally, and support arrives by text to a personal phone.

What it costs

Recurring revenue that was available and never asked for.

What should happen

Offered deliberately at handover, renewed on a trigger.

33%

of integrators name service contracts as their main recurring revenue source. 62% offer them.

Twice as many integrators sell a service contract as actually run one as a revenue line. The difference is fulfilment, not sales.

On CEDIA's own inputs the arithmetic is short. A median firm runs about 30 projects a year at a median residential project price near 20,400 dollars, so roughly 600,000 dollars of installed work. At CEDIA's own 5 to 8% of installed price, a fully attached service programme is worth something in the region of 30,000 to 48,000 dollars of annually recurring revenue, on work that has already been sold.

And it recurs, which is the part that compounds. Three years at that attachment rate on a firm of that size is six figures of recurring revenue requiring no new project, no new client and no additional installation labour. The 29-point gap between offering and running one is where the completion, renewal and handover systems on this page point.

Not our number. Generating Recurring Revenue in the Smart Home Industry, CEDIA, on a survey of more than 1,000 US residential integrators. The project count, project price and contract-share figures are from the same body's market research, which sizes the US professional smart home industry at close to 29 billion dollars across roughly 20,000 integrators.CEDIA is the trade association for this industry. It sells membership, training and certification to integrators, so it has an interest in showing its members an opportunity they are not taking. Two limitations worth stating plainly. The 62% figure describes what firms offer rather than what they collect, and the arithmetic above is arithmetic on published medians rather than a projection for your firm. It is a way of sizing the question, not an answer to it.
Worth doing if

Only start this if youactually want:

Proposals that get an answer

A decision, or the specific reason there is not one yet, instead of silence and a shrinking scope.

Decisions asked for on time

With what they are holding up written into the ask, to a client who did not know.

One packet from sold to installed

So the crew builds what the client actually bought.

Every change written and billed

Priced and approved before the work happens, attached to a payment milestone.

Projects that finish

Punch, programming, training and final payment on a checklist rather than on goodwill.

An installed base worth money

Agreements offered deliberately rather than mentioned at handover.

Support out of your personal phone

Into a queue with context, priority and an owner.

Builders and architects who keep referring

The channel maintained rather than remembered occasionally.

A book you can see

Proposals ageing, decisions overdue and changes unapproved, on one list each morning.

Plainly

We are a build service,not software.

Nothing to log into and nothing to license. We design, build, test and own the systems, and they run inside the platform, inbox and phone system you already pay for.

This is for you if

  • You run custom integration work. Designed systems, real programming scope, real selections. Not retail installs at volume.
  • Several projects run at once. That is the condition that makes coordination the constraint.
  • There is a project management layer. A PM or operations lead, or you are about to hire one.
  • You run D-Tools or a real PM platform. Genuinely in use rather than bought and abandoned.
  • You have an installed base. That is where the recurring revenue is, and it needs history to work from.

This is not for you if

  • You are a solo installer. There is no handoff to protect yet.
  • You are retail or product resale only. Different cycle, different margin, and almost none of the above applies.
  • Your problem is not enough enquiries. Systems multiply a process that exists.
  • You want system design or programming help. That is your expertise. We do not touch it.
Before you apply

What we get askedon the first call.

Do you touch system design or programming?

No. Rack layout, product selection, network architecture and programming are your craft and we have no business in them. What we build is the operating layer around the project: who owns the next contact, what happens when a decision is late, whether the packet matches what was sold, and what you see each morning.

D-Tools already manages proposals and change orders. What is left?

The document and the record are not the problem. What D-Tools does not do on its own is decide who chases a proposal that has gone quiet, branch on whether the delay is the architect or the budget, make sure a change gets written before the crew does the work, and surface the ageing ones on Monday. We build those steps inside D-Tools.

Our clients are high-net-worth and expect a personal service.

Which is exactly why nothing here is written to sound like a system. Messages go out in your voice, approved by you, and anything requiring judgement reaches a person with the project history attached. In practice the client experience improves, because they are told what is needed before it becomes a problem.

Will trade partners have to adopt new software?

No. Builders, architects and trades will not adopt a tool because an integrator asks them to, and any design that depends on it will fail. Where they have to be in the loop we use what they already answer.

What does a build cost?

It depends on what the first system has to do, so we will not put a number on a web page. We scope one at a time and fix the price before anything is built. If we do not think the first one will pay for itself, we will say so on the call.

Show us the proposalthat has been outfor six weeks.

One call. We look at what happens to a proposal, a late decision and a change order on your projects today, what your platform already handles & whether there is a first system worth building.

  • What happens to a proposal today between sending it and hearing back.
  • Where the sold scope and the install packet most often diverge on your projects.
  • What D-Tools or your PM platform already does on its own.
  • How much of your installed base is on an agreement, and how much could be.
  • A straight answer if there is no system worth building yet.

Fourteen questions, about seven minutes. No price, no purchase, and nobody calls you unless you ask them to.

Application1 / 14

Next question: where we send what we prepare.