Management control
Business process audit: what to analyze and how to do it
Many companies grow by adding people, tools and activities without ever stopping to look at how they actually work. For a while, it works. Then the symptoms appear: response times get longer, the same mistakes keep happening, customers get lost between one handoff and the next, and margins shrink for no obvious reason.
At that point the temptation is to buy a solution: new software, one more hire, a consultant for a single department. But if you don't know where the work gets stuck, you risk treating the symptom and leaving the cause untouched. A business process audit exists precisely to find the causes, and to weigh them, before you invest.
In this article I explain what a process audit is, when you need one, which processes to analyze and the method I use step by step, with a practical example with numbers. At the end you will find what you should receive, the most common mistakes and what to do next.
What a business process audit is and when you need one
A process audit is a structured analysis of how the company turns requests into results: how it acquires customers, how it sells, how it delivers the product or service, how it invoices, collects payments and measures. The goal is not to describe the official procedures. It is to understand how things really happen, where they get stuck and how much that costs.
It is different from a certification audit, which checks compliance with a standard, and from a financial audit, which checks the numbers in the accounts. Here the question is practical: where are we losing time, customers and margin, and what is worth fixing first? It is usually part of a broader business and process audit, which also looks at the business model, the offer and margins.
The signs that it is time for a business process analysis are fairly easy to recognize:
- response times to customers are getting longer, even though the team has grown;
- the same mistakes keep repeating: wrong orders, invoices to redo, lost information;
- key people are always overloaded and become a bottleneck;
- revenue is growing but margin is flat or falling;
- you are about to introduce new software, or to hire, and it is not clear what it should solve;
- nobody can say precisely how long it takes to complete a key process.
Which processes to analyze
Not all processes deserve the same level of attention. In an SME it makes sense to start with those that directly affect customers, revenue and margins.
- Acquisition and marketing. Where leads come from, how much they cost, how they are qualified and handed over to sales.
- Sales process. From the moment a request comes in to the moment the contract is signed: timing, steps, conversion rates, who does what.
- Delivery. How the service is produced or delivered, with which handoffs, checks and rework.
- Administration and collections. Quotes, orders, invoices, payment reminders, payment times.
- Data and reporting. Which numbers are collected, by whom, with which tools, and whether they are used to make decisions.
In most cases the audit focuses on two or three of these processes, chosen according to the goal. Analyzing everything in the same depth takes a lot of time and produces few decisions.
The method, step by step
The method I use in process audits has seven steps. The order matters: skipping the first ones almost always leads to the wrong conclusions in the last ones.
1. Define the goal
An audit without a question only produces documents. You start from a specific problem, framed in a measurable way: "quotes go out too late", "margin on projects has dropped", "the team is overloaded and we don't understand why". The goal determines which processes to analyze and in how much depth.
2. Collect the data
CRM, ERP, spreadsheets, analytics, income statements, emails. You collect the data available for the last six to twelve months. It is often incomplete or scattered across different tools: that is also a finding of the audit, because it shows how well the company is able to measure itself.
3. Interview the people
The people who do the work know where time is lost. I run short, one-to-one interviews with concrete questions: what blocks you, what do you do twice, what are you waiting on from others, what information are you most often missing. It pays to talk to people at different levels, because the owner and the people on the ground see the same process in very different ways.
4. Map the as-is processes
You draw the flows as they are today, not as they should be. Who does what, in what order, with which tool, where there are waits. A well-made as-is map makes bottlenecks visible: approvals that depend on a single person, data entered several times, handoffs with no owner.
5. Measure
For each process you measure a few indicators: lead time, waiting time between steps, volumes, error or rework rates, conversion rates. When the data is not there, you work with a sample: for example, you manually reconstruct the last thirty cases.
6. Estimate the impact
For each problem you estimate what it costs: hours lost, customers lost, margin eroded, delayed collections. The estimates don't need to be perfect. They need to be solid enough to compare problems with each other and see which ones weigh the most.
7. Set the priorities
You cross impact with feasibility. The result is a short list, usually three to five actions, in order, with a 90-day roadmap, an owner for each and an indicator to verify the improvement.
A practical example with numbers
A hypothetical example, to make the method concrete. A company that installs technical systems for offices has 25 people, receives about 60 quote requests a month and has an average project value of €8,000. The owner feels they are selling less than they could, but cannot say why.
Goal. Understand why so few quote requests turn into projects.
Data. Reconstructing six months of quotes reveals an average close rate of 20%, or about 12 projects a month. The average time between request and quote being sent is 9 working days. Breaking the data down by band, quotes sent within 3 days close 32% of the time, those sent after more than 7 days only 14%. And half of the quotes, about 30 a month, fall into that last band.
Interviews and mapping. The real flow is: request, site visit, calculation, owner approval, sending. The owner approves every quote above €2,000, which means almost all of them. Breaking down the 9 days: 2 to schedule the site visit, 2 for the calculation, 4 waiting for approval, 1 to send. It also emerges that the price list sits in an outdated spreadsheet and that customer data is entered by hand into three different tools.
Impact estimate. If even just half of the slow quotes, 15 a month, were sent within 3 days with the close rate of the fast quotes, the result would be about 2.7 more projects a month (15 quotes times an 18 percentage point difference). With an average value of €8,000, that is over €20,000 of potential revenue a month. It is an estimate: part of the difference may depend on the type of job, and it needs to be verified in practice. Double data entry, on the other hand, costs about 10 hours a week across three people. It is a real problem, but with a lower impact.
Priorities. The final list is short:
- raise the owner's approval threshold, delegating standard quotes to the technical manager;
- create quote templates for the most frequent jobs and update the price list;
- measure the average time to issue a quote and the close rate by band every week.
The new software to eliminate double data entry is postponed: it makes sense after fixing the process, not before.
What you get at the end of the audit
A well-done audit does not end with a long document nobody reads. It ends with material you can use to decide and to work. At the end you should have:
- the map of the processes analyzed, as they work today, with the critical points highlighted;
- a list of problems with the estimated impact of each, in hours, customers or euros;
- a priority list ordered by impact and feasibility;
- a 90-day roadmap with owners and deadlines;
- a few indicators with their baseline values, to measure improvements;
- an assessment of the quality of the data available and of what is missing to measure better.
The findings are presented in a meeting with the owner and, when it makes sense, with the managers involved. That is when priorities become shared decisions.
Mistakes to avoid
- Starting without a goal. You collect a lot of data and reach few conclusions.
- Analyzing everything in the same depth, instead of focusing on the processes that matter most.
- Trusting the written procedures instead of observing the real work.
- Interviewing only the owner. The view from the top is useful, but problems are easier to see up close.
- Buying software before fixing the process. A messy process, once digitized, stays messy and costs more.
- Ending with a document with no owners, timelines or indicators for the actions.
With clients I also often see a subtler mistake: treating the audit as a judgment of people. If the team thinks it is being evaluated, it hides the problems. It is worth making clear from the start that you are analyzing the process, not the people who work in it.
What to do after the audit
The value of the audit shows in the following months, when priorities turn into concrete actions and the indicators start to move. The natural next step is business process optimization: redesigning the critical flows, clarifying roles and responsibilities, eliminating unnecessary steps and, only at that point, choosing the right tools.
At the same time, it pays to make measurement stable. The indicators defined in the audit need to be updated regularly and read together with the financial numbers: timing, volumes and conversions on one side, costs and margins on the other. If you want to go deeper into the financial side, also read the article on management control for SMEs and growing companies.
After three months it is useful to take stock: what has improved, what hasn't, which new priorities have emerged. Fixing one bottleneck often reveals another one further downstream, and that is normal.
Want to understand where your company's processes get stuck and how much it costs you? See how I work in the business and process audit, or write to me for a first conversation.
FAQ
How long does a business process audit take?
For an SME it usually takes a few weeks, including data collection, interviews, mapping and the presentation of findings. It depends on the number of processes and people involved.
Do you need software to carry out a process audit?
No. You need access to the data, time from the team for the interviews and a method. The tools are chosen afterwards, once it is clear which process to improve.
What is the difference between a process audit and a marketing audit?
A marketing audit analyzes channels, messaging and funnel. A process audit looks at the company's entire operations: sales, delivery, administration and data.
What happens if the data is incomplete?
It is a common situation. You work with a sample, manually reconstructing a number of cases, and the lack of data itself becomes one of the findings of the audit.