Business process improvement is the work of finding the steps that slow a business down, understanding why they happen and redesigning them so work flows with fewer delays, errors and hand-offs. For a growing Nigerian company it is usually the cheapest way to free up capacity, and it should almost always come before a large software purchase.
This guide is for founders, managing directors and operations leads who can feel the business straining as it grows, and want a structured way to fix it.
Signs your processes are holding you back
Most companies don't decide to have poor processes. They grow into them. What worked with twenty people and one location stops working with two hundred people and five. The warning signs are usually familiar:
- Approvals chased on WhatsApp. Purchase requests, discounts and payments wait for a reply from one person, and nobody can see where a request is stuck.
- Month-end runs on spreadsheets. Finance spends days reconciling figures from sales, stores and the bank, because each team keeps its own version.
- The same data is typed in three times. An order is entered in a sales sheet, again in the accounting system and again in a dispatch log.
- Work stops when one person is away. Key steps live in someone's head rather than in a defined process.
- Branches do the same thing differently. Each location has its own way of handling stock, cash or customer complaints, so results are hard to compare.
- Reports arrive too late to act on. By the time management sees last month's numbers, the decisions they should have informed have already been made.
If several of these sound familiar, the problem is rarely effort. People are working hard around processes that no longer fit the business.
Five steps to improving a process
Process improvement doesn't need a large programme to start. It needs a clear sequence. We use the same three stages for every engagement, Define, Transform and Accelerate, and process work fits them in five steps.
1. Map how the work really happens
Start with one process that matters: order-to-cash, procure-to-pay, customer onboarding or month-end close. Walk it with the people who do the work, not just their managers. Record every step, hand-off, approval, system and spreadsheet, and note where work waits.
The goal is the real process, including the workarounds. A map of how things are supposed to work is no use for fixing how they actually work.
2. Find the root causes
For each delay or error, ask why it happens until you reach something you can change. A late payment might trace back to an invoice that needs three signatures, which traces back to a policy written when the company was a tenth of its size. Fix the policy and the delay disappears.
Typical root causes include unclear ownership, approval limits that are too low, missing information at the start of a process, and systems that don't talk to each other.
3. Design the future process
Design the simplest process that still gives the control the business needs. Remove steps that add no value, combine hand-offs, set sensible approval limits and agree who owns each step. Write it down in a form people will use: a one-page process map and a short set of rules, not a hundred-page manual.
Involve the people who will run the new process. They know where the design will break, and they are far more likely to adopt something they helped shape.
4. Decide what to simplify, standardise or automate
Only now ask what technology should do. Some steps should be removed. Some should be standardised across branches. A smaller number should be automated. Automating a broken process only makes the mistakes happen faster.
5. Embed it and measure it
A new process is only real when people use it. Train the teams, update the forms and systems, retire the old spreadsheets and give each process a named owner. Then track a small number of measures that show whether it is working, such as cycle time, error rates, rework and the number of approvals waiting.
Review after the first month and again after the first quarter. Most processes need a round of adjustment once they meet real volumes.
What to automate first, and what to simplify first
A simple test helps decide where automation earns its place. A step is a good candidate for automation when it is:
- stable, meaning the rules don't change every month;
- high-volume, so time saved adds up;
- rule-based, with clear inputs and a predictable output; and
- well defined, with the process already agreed.
Invoice matching, payment reminders, stock reorder alerts, routine approvals within limits and report generation often pass this test. Customer complaints, credit decisions on new customers and anything that changes often usually need simplifying first, and a person in the loop.
Common mistakes
Buying the tool first. A vendor demo can make any process look easy. Without an agreed process and clear requirements, the business ends up bending the software around old habits, or bending itself around the software.
Mapping the ideal instead of the real. If the current-state map doesn't show the WhatsApp approvals and the side spreadsheets, the new design will miss them too, and the workarounds will return.
Leaving out the people who do the work. Changes designed in a boardroom often fail on the shop floor. The people closest to the work see the problems first.
Trying to fix everything at once. Pick the two or three processes that matter most to cash, customers or control, prove the approach, then move on.
No owner after go-live. Processes drift. Without someone accountable for each one, the old ways creep back within months.
When to bring in outside help
Many companies can improve a single process on their own. Outside help is worth it when processes cut across several departments, when leadership can't agree on the future design, when the improvement is tied to a new system, or when the team is simply too busy running the business to step back from it.
A good adviser brings a structured method, an outside view of what good looks like and the time to do the mapping properly. They should also leave your team able to run and improve the processes without them.
At T&A, process work usually sits within a Transformation Blueprint: we map the current state, prioritise the processes that matter most and design the future state, then help implement and embed it. If a new system is part of the answer, see our guide on choosing an ERP in Nigeria.
A quick checklist: is this process ready to automate?
- We have mapped the process as it really runs today.
- We know the root causes of its delays and errors.
- The future process is agreed and written down.
- Each step has a named owner.
- The rules are stable and the inputs are clean.
- We know how we will measure success.
If you can't tick every line, simplify before you automate.
If your business is ready to look at how it runs, talk to us about business transformation. We work with growing companies in manufacturing, retail and other sectors across Nigeria.