Process improvement without an ongoing project
4 min read
A process that has stalled needs a clear diagnosis first. Midsize businesses can examine individual handovers, duplicate data entry or approvals on a small scale and test a change for a limited period. The key is to establish a baseline and a target beforehand. Larger changes, on the other hand, often require a separate project mandate.
At a glance
- Start with the bottleneck. Waiting times, duplicate entries and follow-up questions at handovers provide starting points for a process review.
- Test one change on a limited scale. A team defines responsibility, a timeframe and a metric, then compares the result with the baseline.
- Review controls too. Before removing an approval step or changing how data is transferred, those involved must clarify the effects on quality, access rights and internal policies.
Examine the process before choosing a solution
What is process improvement? Process improvement means systematically examining and changing a workflow to reduce processing time, errors or effort, for example. It can concern a single step or require a larger reorganisation.
A simple overview is enough to start: Where does a case begin? Who handles it? Where does it wait for information or a decision? Employees who carry out the process can explain which exceptions arise and why a step exists.
From this overview, the team can select a specific problem, such as the time between receiving a complete customer enquiry and sending a quotation. The team records how the process currently works and how an improvement could be recognised.
Three starting points in day-to-day operations
Purchasing: If order data is entered into several systems, the team can check whether an existing export combined with a controlled import could eliminate one entry step. Before a test, data quality, access rights and the handling of erroneous records need to be clarified.
Sales: If standard quotations are held up in an approval process, the team can examine whether clear responsibilities and defined limits help. The company decides which quotations still need to be checked based on its risks and internal policies.
Back office: For invoices, mapping the process can reveal which checks actually take place and where cases are held up. An approval stage should only be changed once its purpose is clear and the responsible parties support the new control.
These examples are hypothetical starting points. Only a test in the individual business can show whether a change saves time or causes errors.
Plan, test and evaluate a change
Four weeks is one possible timeframe. This duration is a planning suggestion and depends on how often the process occurs and how significant the risks are. The American Society for Quality’s PDCA cycle describes four repeatable steps for this: plan, test on a small scale, review the results and act on what has been learned.
Plan: The team describes the bottleneck, chooses a change and appoints a person responsible. It measures the existing process over a suitable period and determines before the start what result will count as an improvement.
Test: The change initially runs in a limited area, such as for one product group. Those involved also record errors, follow-up questions and unexpected additional work. A way back to the previous procedure should be agreed.
Review and decide: After the test, the team compares the figures and considers possible side effects. In its guide, the US Environmental Protection Agency, EPA recommends comparing actual process performance with baseline figures and targets. If there are few cases or workloads vary considerably, a longer test may be needed.
If the result is convincing, the team documents the new process and informs those involved. Otherwise, it can adjust or discard the change.
When a project makes sense
A limited test is suitable for a manageable change with clear responsibilities. A larger framework may be needed if several departments are affected, a core system is being replaced or substantial dependencies exist. In such cases, a shared mandate, resource planning and coordinated management help.
Even after a successful test, one question remains: does the change hold up in day-to-day work? A later review of the same metric and feedback from those involved shows whether the new process continues to work.
Frequently asked questions
Which process is suitable for the first test?
A recurring process with a clearly identifiable bottleneck and a limited scope for change is suitable. The people responsible should have time to observe and evaluate it.
Which metric shows an improvement?
That depends on the problem. For quotations waiting to be processed, lead time is a useful measure; for incorrect entries, it is the number of cases returned for correction. The team should also check whether the change has an adverse effect elsewhere.
What happens if the test shows no improvement?
The team first checks whether there are enough comparable cases. It can then adjust the change, test it for longer or return to the previous process.
Read more on MyBusinessFuture
MyBusinessFutureProductivity over cost-cutting programmesMyBusinessFutureWhen AI works alongside you instead of assistingMyBusinessFutureThe partnership that carries more than any acquisitionMore from the MBF Media Network
Digital ChiefsWhich IT budget will survive the spending cutscloudmagazinThe inference bill nobody budgeted forSecurityTodayWhere midsize businesses still lag on NIS2 technical requirementsImage source: AI-generated (November 2026)
Translated from the German original using artificial intelligence. The German version is authoritative.
