Branditify

ERP & operations systems

Run the job once. Let every team see the same truth.

We build custom ERP and operations systems that connect the records, handoffs and approvals your teams share — so one job has one owner, one status and one history instead of five.

Follow one job through
  • Sales CRMApprovedClosed and handed over
  • Planning sheetWaitingRow 214, last edited Friday
  • Team chatStock confirmedA message, three days ago
  • Operations boardNot startedNobody moved the card
  • AccountsNo invoice yetWaiting to hear it is done

JOB #4812 — same job, 5 different answers.

JOB #4812Complete

Northline Interiors · 12 custom fixtures

Owner
Operations
Stage
Complete
Waiting on
Finance to raise the invoice
Next action
Invoice the job
History
  1. Order confirmed by sales
  2. Planned for the week of the 8th
  3. Requirement checked — 4 units short
  4. Purchase request approved
  5. Material received and issued
  6. Work started
  7. Sent for quality check
  8. Signed off — 12 of 12
  9. Invoiced against the job

One job. It already exists in five places.

Illustrative record · sample data

Where it starts

ERP stops being jargon the moment you follow one job.

Not modules. Not departments. One piece of work, and what has to happen to it.

A custom ERP system connects the records and workflows several parts of a business need to share — orders and jobs, stock and purchasing, approvals, operations and the commercial side. What it should contain is decided by how the business actually works, which is why the useful conversation starts with one real job rather than with a list of modules.

  1. The customer says yesA confirmed order for 12 custom fixtures
  2. What has to happen next?Somebody has to plan the work
  3. Who owns it?Operations — and that has to be recorded, not assumed
  4. What do they need?What was agreed, by when, and whether the material exists
  5. What has to be true before the next team starts?The material is confirmed and the plan is set

Answer those five for one job and you have described the system. Everything after that is scope.

One record

Move the job once. Watch what becomes true everywhere else.

This is the part every ERP page describes as “real-time” and never shows.

In a connected operations system there is one record for the job and every team reads the same one. Sales sees where it has got to, operations sees what to do next, procurement sees what it needs, finance sees whether it can be invoiced yet, and management sees whether it is moving. Nobody is updating anybody: the state changes once and the consequences are already true.

Move the job through its stages.

JOB #4812In production

Northline Interiors · 12 custom fixtures

Owner
Operations
Stage
In production
Waiting on
Nothing — work is running
Next action
Quality check
History
  1. Order confirmed by sales
  2. Planned for the week of the 8th
  3. Requirement checked — 4 units short
  4. Purchase request approved
  5. Material received and issued
  6. Work started
  7. Sent for quality check
  8. Signed off — 12 of 12
  9. Invoiced against the job
What that means right now
  • SalesIn production
  • OperationsRunning · 6 of 12 complete
  • ProcurementDelivered and issued to the job
  • FinanceNot invoiceable yet
  • ManagementActive · on schedule

Finance did not become invoiceable because somebody sent a message. It became invoiceable because the work was signed off.

Handoffs

The moment work changes hands is where operations actually leak.

Most operational failures are not bad work. They are a handoff nobody owned.

A good operations system makes handoffs explicit. When a stage finishes, the next team should know the work is ready, what information came with it, what is still missing, and who owns the next action — without anybody having to send a message or remember to check.

How it usually happens
  • A message in a group chat
  • A row edited in a shared sheet
  • A verbal “it’s done”
  • Nothing — and it sits for two days
What the system does instead
  1. Stage marked complete
  2. Required checks pass
  3. Next owner assigned
  4. It appears in that team’s queue
At the handoff, the job says
Owner
Operations
Stage
Planned
Waiting on
Material to be confirmed
Next action
Check what the job needs
Who owns this now?
One name, on the record, not in somebody’s head
What is it waiting on?
Named, and visible to everyone who cares
What is blocking it?
Recorded as a reason, with a next action

Rules

Business rules can live in the workflow instead of in somebody’s memory.

The rule that only one person knows is the rule that gets broken on a busy week.

Approvals belong inside the workflow: the action is simply not available until the approval is given, and who approved it is recorded on the job. That is more reliable than a policy people are expected to remember, and it means the history answers the question later without anybody reconstructing it.

Purchase request · 4 units of trimWaiting for approval
For
JOB #4812
Requested by
Operations
Reason
Four units short of the requirement
Needs
Manager approval

Approved by R. Menon · Tue 09:40

What approval unlocks
  • The purchase action becomes available
  • The job’s dependency is recorded
  • Finance can see the committed cost

Which actions need approval, and from whom, is decided per business. Some workflows have none.

What the job needs

A job is only as ready as the thing it is short of.

Stock, or people, or a machine, or a slot in the calendar. Same shape.

Connected operations means the job knows what it requires and whether that exists. When something is short, the shortfall becomes a request with an expected date, and the job that depends on it shows the dependency rather than quietly slipping. The same structure works whether the constrained thing is material, a person’s time or a piece of equipment.

A product businessBrushed trim · 40mm

8 of 124 short

  1. Purchase request raised
  2. Expected 11 Sep
  3. JOB #4812 shows the dependency
A service businessSenior installer · week of the 8th

1 of 21 short

  1. Resource request raised
  2. Alternative slot proposed
  3. JOB #4812 shows the dependency

The page shows both because this is not a manufacturing-only idea. What a job depends on differs; the fact that it should be visible does not.

When it does not go to plan

An operations system earns its place on the bad week, not the good one.

Anything can run a job that goes perfectly. The question is what happens when it does not.

When something goes wrong the job should not go quiet. A blocked job names the reason, shows what it puts at risk, has an owner and a next action, and stays visible in the places people already look. The purpose is not to prevent problems — it is to make sure a problem is never a surprise a week later.

Material ETA movedBlocked
Reason
Supplier pushed delivery to the 13th
Impact
Production date at risk
Next action
Review the revised schedule
Owner
Operations
Quality check failedBlocked
Reason
Two units rejected at sign-off
Impact
Two units to remake
Next action
Re-plan the shortfall
Owner
Operations
Approval not givenBlocked
Reason
Purchase above the agreed threshold
Impact
Material not ordered
Next action
Escalate to the manager
Owner
Procurement
Customer changed the orderBlocked
Reason
Two fixtures changed after approval
Impact
Plan and cost both move
Next action
Re-confirm scope and price
Owner
Sales

Every one of those is a normal week in an operations business. None of them should require somebody to notice.

How much of it

An ERP does not have to mean every department on day one.

The version that works starts where the friction actually is.

No. A useful ERP can begin with the operational records and handoffs causing the most friction — usually the job itself, its owner, its stages, its approvals and its history — and expand once the connected model has proved useful. Starting with every department at once is how ERP projects become the thing everyone has a story about.

  1. Where it startsThe operational spine, almost always
    • Jobs and orders
    • Owners and stages
    • Approvals
    • History
  2. Connect nextOnce the spine is being used
    • Stock and procurement
    • Invoicing and costs
    • Operational reporting
  3. Later, if it earns itDecided by evidence, not by the module list
    • HR and people
    • CRM and pipeline
    • Customer or vendor portals
    • Mobile
    • Advanced automation

Some of those later items are already Branditify products. Connecting one is often cheaper than building it into the ERP.

Two decisions

Before anyone builds anything.

Whether it should be custom at all, and whether what you want is actually an ERP.

Custom is not automatically the right answer, and plenty of businesses are better served by configuring something that already exists. Equally, a lot of what gets asked for as an ERP is really a CRM, and the two solve different problems: a CRM owns the relationship and the deal, and an operations system picks the job up once the customer has said yes.

Buy it when your workflow is ordinary. Build it when it is not.

Existing ERP software is probably better when
  • The workflow is standard for your industry
  • A product already covers most of it
  • Configuration gets you the rest
  • The business can adapt to the product
A custom system makes sense when
  • The operational workflow is materially different
  • Several systems create duplicate work
  • The business rules are specific to you
  • Teams need one connected record
  • Control, integration and ownership matter

A CRM owns the deal. An operations system owns what happens after it.

A CRM owns
  • Leads and enquiries
  • The pipeline and the deal
  • Quotes and the relationship
  • Everything up to “yes”
An operations system owns
  • The job that “yes” creates
  • Planning, materials, execution
  • Handoffs between teams
  • Completion, invoicing and history

What you already run

Nobody is asking you to throw everything away.

Most of what a business runs is doing at least one job well. The question is which.

We review what each current system already does well before deciding what should stay, connect or move. Accounting is usually kept and connected rather than rebuilt. A spreadsheet that is really a workflow usually moves into the system. A legacy database is often kept as a source while the new records take over. What can actually be connected depends on what each system exposes, which we confirm before designing around it.

  • KeepIt works. Leave it alone.
  • ConnectIt stays, and the systems talk.
  • ReplaceIt was never really software.
  • MigrateThe records move across.
  • AccountingKeepKept, and connected where it can be
  • CRMConnectKeeps the pipeline; hands the job over
  • Planning spreadsheetReplaceThis one is really a workflow
  • Legacy databaseMigrateRecords move; the old system can stay readable
  • Website / order formsConnectBecomes a way jobs arrive
  • Warehouse or shop-floor toolsConnectWhere they expose anything to connect to
And the old records
  1. Sample it
  2. Map the fields
  3. Clean what is broken
  4. Import
  5. Verify against the source
  6. Go live

Migration is judged on what the business needs to keep working, not on moving every historical field perfectly. Some old data is worth carrying across in full, some is worth summarising, and some is best left readable in the old system. We decide that per source, with you, before importing anything.

Visibility

Reporting is trustworthy when it comes out of the work itself.

Not a dashboard bolted on afterwards. The same records, counted.

Operational reporting is a consequence of the records being connected, not a separate product. If every job has a real stage, a real owner and a real history, then what is active, what is blocked, what is waiting for approval and what is due this week are simply counts of that — and they are as current as the work, because they are the work.

  1. 18Active jobs
  2. 3Blocked
  3. 2Waiting approval
  4. 7Due this week

Every one of those numbers opens to the jobs behind it — including JOB #4812.

Illustrative counts from the sample data on this page.

Dashboards When the reporting itself is the project

Scope

What makes one operations system much larger than another.

Rarely the industry. Nearly always the count.

Scope on an operations build is mostly arithmetic: how many workflows, how many teams and roles, how many shared records, how many approvals and permissions, how many other systems it has to talk to, and how much history has to move. What affects development time most is the number of distinct things to design, build, test and support — plus the state of the data being carried over and how many decisions are still open when the build starts.

What changes the size of a build
  • WorkflowsHow many distinct operational processes
  • Teams and rolesHow many kinds of person use it
  • Shared recordsJobs, items, customers, suppliers, and how they relate
  • IntegrationsEach other system is its own piece of work
  • MigrationHow much history moves, and what state it is in
  • ApprovalsHow many rules, and who holds them
  • PermissionsWho can see and change what
  • Stock or resourcesWhether the system tracks what jobs consume
  • ProcurementWhether buying happens inside the system
  • BillingHow far into the commercial side it goes
  • PortalsWhether customers or suppliers get their own access
  • Branches and entitiesSeparate locations, companies or books
  • MobileWhether the work happens away from a desk
  • ReportingWhat has to be counted and shown
  • Audit and historyHow much has to be recorded and kept
  • ComplianceWhere a standard genuinely applies
Branches, warehouses and multiple companies

A business running several locations, warehouses or legal entities usually needs separate permissions, its own numbering, and reporting that can be read per entity and together. It is worth deciding early because it shapes the data model — but it is not something every project needs, and adding it by default is a common way to make a first release twice the size it had to be.

Where it runs

Cloud or on your own infrastructure, decided by what the business already runs, what it has to integrate with, and any requirements that apply to your data. We scope that with you rather than defaulting to one answer.

Access and permissions

Roles decide what a person can see and do, sensitive fields can be restricted further, high-impact actions can require approval, and what happened is recorded on the record. Where a specific standard applies to your industry or data, we confirm what it requires and design the scope around it.

Who owns what

The system, its code and its data are yours. Work is handed over with the repository, the deployment and the access it runs on, and with documentation where that is in scope. Hosting, maintenance and support are arranged around the team you have — some clients take it all in-house at handover, some keep us on. What applies to a project is written into that project’s scope rather than assumed here.

Questions

ERP and operations systems, answered.

What is an ERP system?
An ERP system is software several parts of a business share, so that the same records — orders and jobs, stock, purchasing, approvals, costs — are read and updated in one place instead of separately by each team.
What is a custom ERP system?
One built around how your business actually works rather than configured from a product’s assumptions. It connects the records and workflows your teams genuinely share, and leaves out the parts of a standard ERP you would never use.
What is the difference between an ERP and operations software?
In practice, very little — “ERP” usually implies a wider scope covering finance and procurement as well as operations. Both mean the same thing that matters: teams working from connected records instead of separate tools.
What is the difference between an ERP and a CRM?
A CRM owns the relationship and the deal — leads, pipeline, quotes, everything up to the customer saying yes. An operations system owns the job that “yes” creates: planning it, resourcing it, doing it, handing it between teams, completing and invoicing it.
Custom ERP or off-the-shelf ERP?
Off-the-shelf is usually better when your workflow is standard for your industry, an existing product covers most of it, and configuration gets you the rest. Custom makes sense when the operational workflow is materially different, several systems are creating duplicate work, or the business rules are specific enough that you would be fighting the product.
Does every business need an ERP?
No. Plenty of businesses run well on a few good tools. The signal that something connected is needed is usually duplication — the same job being re-entered, re-explained or re-checked as it crosses teams.
Can an ERP start with one process?
That is usually the right way to start. Begin with the operational spine — the job, its owner, its stages, its approvals and its history — and expand once that is genuinely being used.
Does an ERP need every department?
No. A first release covering the operational records and handoffs causing the most friction is more useful than a system covering every department shallowly, and far more likely to be adopted.
Can it connect to our existing CRM, accounting or other software?
Where those systems allow it. We confirm what each one can expose — what can be read, what can be written, what it will not permit — and design around what is actually available rather than assuming an integration exists.
Do we have to replace the systems we already use?
Usually not. Accounting is normally kept and connected. A CRM keeps the pipeline and hands the job over. What tends to get replaced is a spreadsheet that was doing a workflow’s job all along.
Can our spreadsheet or legacy data be migrated?
Usually, with judgement about what is worth moving. We sample the source, map the fields, clean what is broken, import and verify against the original. Some history is worth carrying in full, some is worth summarising, and some is best left readable in the old system.
Can different teams have different permissions?
Yes. Roles decide what each person can see and do, sensitive fields can be restricted further, and high-impact actions can be put behind an approval.
Can approvals be built into the workflow?
Yes, and that is usually better than a policy people are expected to remember. The action is unavailable until approval is given, and who approved it is recorded on the record.
Can exceptions and escalations be tracked?
Yes. A blocked job can name its reason, show what it puts at risk, carry an owner and a next action, and stay visible where people already look, rather than going quiet until somebody notices.
Can it include inventory and procurement?
Yes, where the business needs it. A job can know what it requires, whether that exists, and what to do when it is short — including raising a purchase request and showing the resulting dependency on the job.
Can it include customer or supplier portals?
Yes, as a later stage rather than a first release. Once the internal records are trustworthy, giving a customer or supplier a view of the parts that concern them is a much smaller piece of work.
Do we need a mobile app?
Only if the work happens away from a desk. A responsive web system covers most office and back-office use, and is cheaper to change. Shop floor, warehouse or field work is where a mobile experience genuinely earns its place.
Who owns the system, code and data?
They are yours. Work is handed over with the repository, the deployment and the access it runs on, and with documentation where that is in scope. What applies to a specific project is written into that project’s scope.
What determines the scope of an ERP project?
How many workflows, how many teams and roles, how many shared records and how they relate, how many approvals and permissions, how many systems it must talk to, and how much history has to move.
What makes one ERP project much larger than another?
Usually the count rather than the industry: more workflows, more roles, more integrations, more migration. Multiple branches or legal entities is the single change that most often doubles a project, because it reaches into permissions, numbering and reporting at once.
What affects development time?
The number of distinct things to design, build, test and support, the state of the data being carried over, and how many operational decisions are still open when the build starts. We scope from your actual workflows rather than quoting a standard number.
What happens after launch?
The first weeks of real use show which stages people actually work through, where work stalls, and what the team is still doing outside the system. That decides what gets built next, and it is a better guide than anything decided in advance.
Can the system expand later?
Yes, and it is designed expecting to. The records and permissions are shaped so that adding a workflow, a team or a connected system later does not mean rebuilding the middle of it.

Start here

Bring us the job that keeps crossing too many tools.

Not a requirements document. One piece of work, and everywhere it currently lives.

What starts the job
An order, an enquiry, a contract, a date
Which teams touch it
And in what order, on a normal week
Where the information lives now
The tools, the sheets, the chats
What has to be approved
And by whom
Where it gets stuck
The part everyone already complains about