Undocumented processes
Each person does it their way, so quality depends on who is on shift.
Processes, controls and automation
Software only automates the process you give it. If the process is unclear, undocumented and dependent on individuals, the system simply records the confusion faster. We look at how work actually moves through the business, remove the steps that add nothing, put controls where risk sits, and then automate what is left.
Each person does it their way, so quality depends on who is on shift.
Everything routes to one person, so work waits and control becomes a formality of rubber-stamping.
Data re-keyed between systems, emails and spreadsheets, with no audit trail.
No segregation of duties, no approval limits, no reconciliation discipline.
Month-end measured in weeks because the process is sequential and manual.
The most expensive mistake available: automating a process that should not exist.
Process work is scoped by process or by function. Typical deliverables below.
Documenting how work currently moves — order to cash, procure to pay, stock, payroll, expense, project billing — with handoffs, delays, rework and decision points identified. Output: as-is process maps with pain points marked and time or cost estimates where measurable.
Designing who can approve what, at what value, with what evidence, and what happens when they are unavailable. Output: a delegation of authority matrix.
Controls placed against the actual risks: segregation of duties, matching, reconciliations, access rights, physical controls and system-enforced rules. Output: a control matrix linking each risk to a control, its owner and how it is tested.
Written procedures for the processes that matter, at a level of detail a new starter can follow. Output: an SOP set, version-controlled and owned.
Automating recurring journals, bank matching, invoice capture and coding, approval routing, reminders and reconciliation, using the capabilities of the platform in use. Output: configured automations with exception handling and a monitoring routine.
Rebuilding month-end as a parallel, scheduled process with a task list, owners and deadlines. Output: a close calendar and checklist with a target close date.
Reviewing user roles, permissions and approval configuration in the accounting or ERP system so that the system enforces the controls rather than relying on discipline. Output: a role and permission design with change recommendations.
Supporting the change: training, communication, a trial period, measurement of the before and after, and adjustment. Output: an adoption review with measured results.
We start with the one or two processes causing the most pain or carrying the most risk, not everything at once.
We watch the work being done and talk to the people doing it. What management believes happens and what happens are frequently different, and both matter.
Waste, delay, rework, risk and cost are identified and, where possible, quantified so the case for change is clear.
With the people who will run it. A process designed without them will not survive contact with reality.
SOPs, approval matrix and control matrix, so the improvement does not decay when attention moves elsewhere.
Only now. The remaining process is configured in the system, with exceptions handled rather than ignored.
We measure against the baseline and adjust. If the improvement cannot be measured, it did not happen.
Preparation is where most engagements are won or lost. The more of this you can gather before we start, the faster the work goes and the more accurately we can scope it.
Improvement depends on adoption. We measure the baseline, support the change and review the result, but we cannot promise a specific reduction in cost or time before understanding the process.
We do not publish a price list. The drivers below vary too much between businesses for a published figure to be honest — and a price quoted before an assessment is usually wrong in one direction or the other.
| Factor | How it affects the engagement |
|---|---|
| Number of processes in scope | We recommend starting with one or two rather than a whole business review. |
| Number of sites and teams | Each location may run the process differently, and all versions must be mapped. |
| System capability | What can be enforced in the system versus what must remain a manual control changes the design and the effort. |
| Change appetite | Where approval authority has to move away from the owner, the change management effort is as significant as the design. |
| Documentation required | A control matrix for a funder or auditor is more formal than an internal SOP set. |
Process engagements are quoted per process or per phase after an initial assessment. Automation configuration may be quoted separately where it involves development work.
Often not. A surprising amount of improvement comes from removing steps, clarifying approvals and enforcing controls in the system you already have. Where a system change is genuinely needed we will say so, and where it is not we will say that too.
A single process review is typically a matter of weeks including observation, mapping, design and documentation. Whole-business reviews take longer and are usually phased. We scope against the specific processes you select.
By involving them in the design. The staff doing the work usually know exactly what is wrong; they are rarely asked. We also start small, prove the improvement on one process, and let the result make the case for the next.
A document that sets out who may approve what, at what value, with what supporting evidence, and who acts in their absence. It is one of the simplest and most effective controls a growing business can put in place.
Yes. Where documentation is required for a specific external party, we scope to their standard and format, and we can support the questions that follow.
With ownership, a control matrix that is tested periodically, and system-enforced rules wherever possible. Controls that depend on memory decay; controls that the system enforces do not.
Tell us what you are dealing with. We will tell you honestly whether we can help, what it would involve and what it would cost.