Anna Totterdell
Projects Director
Twelve weeks. That is what it takes to see measurable, operational results - without replacing your ERP, without migrating your data to a new platform, and without launching a transformation programme that consumes your leadership team for a year.
I am not talking about a proof of concept. I am not talking about a strategy document. I am talking about live, working improvements in how your business operates - with before-and-after metrics that you can put in front of your board.
Here is how.
The premise
Most mid-market businesses do not need new systems. They need their existing systems to work together. They need the manual handoffs automated. They need the data structured. They need the processes that currently depend on individuals to be captured in workflows that anyone can follow.
This work does not require replacing anything. It requires data and systems integration across what you already have, structuring the data that flows between them, and automating the repetitive, manual steps that consume your team's time.
And it can be done in phases - with each phase delivering measurable value and building the foundation for the next.
Weeks 1–3: Diagnosis and scope
The first three weeks answer a single question: where is the highest-value work?
This is not a generic assessment of the entire business. It is a targeted investigation of the specific workflows where time, cost, and risk are concentrated - the processes that leadership already knows are painful but has never dissected properly.
For each priority workflow, we establish three things. First, how the process actually runs today - every step, every system, every point where someone intervenes manually. Second, the state of the data that flows through it - where it originates, where it conflicts, where it breaks down. Third, the integration gaps - which systems are involved and how they currently exchange information (or fail to).
By the end of week three, you have a scoped plan: which workflows are being fixed, what the expected improvements are, and what the before metrics look like so you can prove the change.
Weeks 4–8: Implementation
This is where the operation changes. The work is intensive but tightly scoped - you are not rebuilding the business, you are fixing the specific workflows identified in the first three weeks.
The integration layers go in. Systems that were exchanging data via email or spreadsheet now exchange it automatically, in real time, with the correct structure. The data that feeds these workflows gets normalised - consistent formats, deduplicated records, agreed definitions - so that every system is working from the same truth.
The manual steps get replaced with business automation at the workflow level. Not individual tasks - complete workflows, from trigger to completion. The approval that used to wait in someone's inbox is now routed, escalated, and tracked. The status report that took a day to compile is generated live from connected data.
And the visibility gets built. Leadership dashboards that show how the fixed processes are performing - not based on last month's data, but on today's. Cycle time, throughput, error rate, cost per transaction. The numbers that matter, updated without anyone having to assemble them.
Weeks 9–12: Prove and compound
The final phase turns working systems into proven results.
The automated workflows run in production. Edge cases that did not surface in testing get caught and handled. The team transitions from the old way of working to the new one - with support, not a hard cutover.
Then the measurement. Compare the current performance against the baselines from week one. The numbers are not estimated - they are taken from live operation. Hours recovered. Errors eliminated. Cycle times reduced. Cost per transaction changed. These are the figures that go in front of the board.
And then the compounding starts. The integrations from this cycle extend naturally to adjacent workflows. The data standards apply across more of the business. The automation patterns replicate. The second twelve-week cycle delivers more than the first, because the groundwork is already in place.
This is how operational improvement actually scales - not through a grand programme, but through repeated, measurable cycles that build on each other.
What this looks like in practice
Let me give you a concrete example.
A mid-market business with an ERP, a CRM, and a separate finance system. Their order-to-cash process involves sales confirming orders in the CRM, operations re-entering them in the ERP, finance manually creating invoices based on a spreadsheet extract, and a monthly reconciliation that takes the finance team three days.
In twelve weeks:
The CRM and ERP are connected. When a sales order is confirmed, the data flows automatically into the ERP without re-entry.
The ERP and finance system are connected. Invoices are generated automatically based on order completion, with the correct data, in the correct format.
The monthly reconciliation that took three days is replaced by a live dashboard that shows the current state in real time. Discrepancies are flagged automatically as they occur, not discovered three weeks later.
The result: the finance team gets three days back every month. Manual re-entry errors drop to zero. The sales-to-invoice cycle shortens by 60%. And leadership has real-time visibility into cash flow for the first time.
No new systems were purchased. No ERP was replaced. No data was migrated. The existing tools were connected, the data was structured, and the process was automated.
Why twelve weeks matters
Twelve weeks is long enough to deliver real, substantive operational change. It is short enough to maintain executive attention and team momentum. And it is measurable enough to prove the value before anyone has to make a larger commitment.
Most mid-market businesses have been told - usually by tech consultancy that benefits from long programmes - that operational improvement takes twelve to eighteen months. That is true if you are replacing systems. It is not true if you are connecting them.
The work is not theoretical. The results are not estimated. And the approach does not require you to bet the business on a technology decision that takes two years to validate.
Twelve weeks. Measurable results. No system replacement. That is the proposition.
