The company knows its current way of working is no longer enough. Information travels through emails and spreadsheets, the team keeps asking about statuses, and every change needs manual coordination. A decision follows: we need a system, or we need to extend the one we have.
The next step is usually a list of features. Yet system development in a company is easier to start from a single process: describe how it runs, where it gets stuck, who owns it and what result is expected. The first scope of work follows from that description.
A feature list is not a scope yet
The enquiries we receive usually take the form of a long list: customer database, calendar, employee panel, reports, notifications. Each item is easy to understand. The list does not say, however, which of them solves the most painful problem, or in what order they should be built.
A feature answers the question "what should the system be able to do". A process answers the question "what should happen in the company, who is responsible, and how will we know it worked". Without the second answer it is hard to compare offers, and even harder to judge whether the implementation changed anything.
Choose one process, not the whole system
One process is enough to begin with. A good candidate is a place where work regularly stops: someone waits for information, copies data from one tool to another, or makes a decision without a current picture of the situation.
It does not have to be the organisation's biggest problem. What matters is that you can point to where the process starts and ends, and to the person responsible for it. If the company already uses a system that has grown over the years, first establish whether the limitation lies in the process or in the solution itself. We write about this in The system still works, but it can no longer keep up. Improve, rebuild or replace?
Describe the flow: input, people, decisions, data, result
Describing a process does not require a specialist notation. Five questions are enough:
- Input: what triggers the process? An order, a request, a change of plan?
- People: who takes part and who is responsible for the result?
- Decisions: at which points does someone have to approve or choose something?
- Data: what information is needed and where is it today?
- Result: what is the outcome and who uses it?
Such a description gives everyone a shared language. The Lean Enterprise Institute says of value-stream mapping that it provides a common language for talking about a process and shows the linkage between material flow and information flow. System design is about the same thing: before a screen is built, you need to know what information should appear on it and who needs it.
Example: order status in a manufacturing company
The example below is hypothetical. It illustrates the questions and does not describe a specific client.
A manufacturing company struggles to establish the current status of its orders. A salesperson asks the production manager, the manager checks a board on the shop floor and a spreadsheet, and the customer waits for an answer. Instead of starting from an idea for a "status module", it is worth asking:
- Who records a status change, and at what moment?
- Where is that information stored, and is there one place everyone trusts?
- Who uses it: production, sales, the customer?
- When is the information needed, and what happens when it is missing?
The answers may show that the problem is not a missing status screen, but a missing moment at which someone updates the status. That changes the scope of the first stage.
The first process card
The information you collect fits on one page:
- Process name.
- Owner: the person responsible for the result.
- Current flow: steps from input to result.
- Limitation: the place where work stops.
- Business impact: what the company loses or risks.
- Available data: what information exists and where.
- Expected result: what should work differently.
- Dependencies: other systems, departments and integrations.
- Questions to explore: what we do not know yet.
The card helps you gather information before talking to a supplier. It does not replace analysis or Discovery, but it shortens the way to them.
How you will recognise the improvement
Before work starts, write down how you will recognise the change. It may be the time from request to answer, the number of status questions, or the number of manual corrections in the data. Measure the current state first, even roughly.
Declaring savings before measuring gives you little. A reference point lets you check after implementation whether the first stage met its goal, and decide about the next one on that basis.
From the first process to the first scope
A description of one process is enough to start a factual conversation. Companies sometimes write to us directly: at this stage we are not asking for a quote, we want to understand how the cooperation works. That is a good starting point. You can also read about connecting processes across departments in Business operating system: how to connect company processes?
Book a free consultation. During the conversation, which takes up to 60 minutes, we get to know the organisation, the current tools and the main problem. After the meeting you receive a short summary and a proposed next step: a quote for a smaller project with a clear scope, paid Discovery before quoting a large project, or an honest statement that we are not the right partner.
How do you rate this article?
Your rating:
average rating 5 / 5. number of ratings 1