Every brief has a point of view
A business may have capable people responsible for sales, operations and finance. Information nevertheless flows through different tools, personal conversations and decisions taken by the owner. The commissioned system will have its own defined purpose and interface with only some of that activity.
The person who requests software can describe the work within their experience and authority. Other functions may have information they cannot see, and decisions they cannot make. The agreed specification can be sound without capturing those other requirements. Acceptance against that specification does not establish what the company may need next.
The connection to Conway's Law
Conway's Law describes a tendency for systems to reflect the communication structures of the organisations designing them. It does not say that every system mirrors an org chart, or that all departmental boundaries are harmful.
For a developer, a related mechanism operates at requirements time: who participates, what each person sees and what each can authorise shapes the brief. This is a useful inference from organisational boundaries, not a direct prediction made by Conway’s Law. It does not establish that the omitted work is a problem.
Why the original scope matters
An operations team may sponsor a system to support its own work. Other decisions can remain with the owner, finance or customer-facing staff. Those decisions were not necessarily omitted by mistake: the project may have had sound reasons to exclude them.
A later discussion should not begin by assuming that the business needs another application. It should establish whether a different activity matters to the client, whether the current arrangements are adequate and who has the authority to consider a change.
Why this is not an upsell exercise
Product-led upselling begins with a service the supplier can offer. Business research begins with what the client sells, delivers, promises and needs to decide. These routes can produce very different results. A plausible change may call for a different policy, a clearer owner, a modest data register, substantial software development or no intervention.
The developer can then discuss the relevant business question with the client. The client decides whether it matters. Only if there is a validated need does normal requirements discovery begin.
What an initial outside-in check can and cannot do
Public information may describe services, contracts or recurring commitments. A developer can describe the functions and records in the original scope without sharing source code or confidential records. Comparing those two descriptions may identify a decision the original application did not address.
They cannot prove which capabilities the company lacks, what its internal data shows, who would approve a change, or whether the client will buy another project. If the evidence is insufficient, the correct conclusion is to stop or request authorised discovery — not to manufacture a problem.
That is the purpose of the £50 Account Screen: to assess one existing client and decide whether there is a defensible next question.
Spreadsheets are another view of that same boundary. A client's Excel files can clarify established work without establishing every other business need.