Making a Fictional Customer Enquiry for a Software Prototype Review

Prepared for Blackstone Intelligence.

A practical worksheet for Malaysian service businesses to write a plausible, entirely fictional customer enquiry and use it to walk through a software prototype review, recording what happens when a step has no clear outcome.

Why a Fictional Enquiry Beats a Demo Script

Most prototype walkthroughs drift toward the happy path. Someone clicks the button that works, the screen changes, and the room nods. The awkward parts, such as the enquiry that arrives with two addresses, a name spelled three ways, or a request for something the system has no field for, never get typed in.

A fictional customer enquiry fixes that. You write one short, invented message before the review, then use it as the only input for the whole session. Because it is fictional, nobody has to worry about exposing a real customer's details, and you can make it as awkward as you like on purpose. The goal is not to test the software's speed or to prove it works. The goal is to see whether the team can trace one enquiry from arrival to a clear outcome, and to write down every point where the trace breaks.

This guide is written for Malaysian service businesses that maintain a public website and are reviewing a prototype, whether that prototype is a small internal tool, a customer enquiry flow, or part of a larger build. Blackstone Intelligence, a technology consultancy based in Kuching, Sarawak and operated by Blackstone Consultancy Sdn Bhd, lists software development and AI consulting among its services, so a review like this is the kind of groundwork that often precedes a build conversation. The worksheet below is a writing aid, not a test plan, and it makes no claims about how any particular system will perform.

The Worksheet: Seven Fields, One Enquiry

Write the enquiry as a single short message, the way a stranger might send it through a website form, a WhatsApp message or an email. Keep it under about 150 words. Fill in each field below with invented details only, and mark anything you are unsure about as a question for the reviewer rather than guessing.

Field 1, sender name. Use an obviously fictional name, such as a common first name paired with a placeholder surname. Note how the name is written: all capitals, with a title, with a hyphen. This field maps to the name field in the enquiry record and to any greeting the system generates.

Field 2, contact channel. Decide whether the enquiry arrives by form, messaging app or email, and write the message in the style of that channel. This maps to the intake step and to whatever routing rule sends the enquiry onward.

Field 3, location. Use a Malaysian town or city that is plausible for your service area, and add one detail that complicates it, such as a customer who lives in one place but wants the service at another. This maps to any location or coverage field and to how the system handles two addresses.

Field 4, what they want. Describe the request in the customer's own words, not your internal service names. Include one vague phrase, such as asking for a quote without saying for what. This maps to service selection, categorisation or free-text description.

Field 5, timing. State a preferred date or a rough window, and make it slightly ambiguous, for example next month or after the holidays. This maps to scheduling fields and to how the system treats an unclear date.

Field 6, one complication. Add a single realistic wrinkle: a follow-up question, a request to be contacted only in the evening, or a mention of a previous enquiry. This maps to notes, preferences or duplicate handling.

Field 7, what a good outcome looks like. Write one sentence describing what should happen to this enquiry, such as a reply within a stated period and a record that can be found later. This is your reference point for the whole review.

Keep the worksheet in a shared document with the date and the names of the people reviewing. Do not include real customer information, staff personal details, passwords, system credentials or anything marked confidential.

Mapping Each Field to a Task in the Walkthrough

Before the session, turn each field into a task and write the task in plain language. A task is a single question the reviewer must be able to answer by looking at the prototype. For example: can the reviewer find the sender's name after the enquiry is submitted? Can they see which channel it came from? Can they tell whether the location was understood?

Order the tasks the way the enquiry would travel: arrival, reading, categorisation, response, record. For each task, write down what you expect to see and where you expect to see it. This expectation is what makes the review useful, because a missing expectation is easy to mistake for a working screen.

Assign one person to operate the prototype and one person to record. The recorder should not be the same person who built the screen, because the builder already knows where everything is. If the team is small, invite someone from a different part of the business, such as the person who normally answers enquiries, to sit in as recorder.

Set a time box for each task. Fifteen to twenty minutes per task is a reasonable starting point for a small prototype, but treat that as an example rather than a rule. If a task runs long, stop and record why rather than pushing through, because the delay itself is a finding.

Where the prototype is part of a wider build, it helps to know what the surrounding services are meant to cover. Blackstone Intelligence's published terms list AI automation, workflow automation, software development, integrations and training among its services, which gives a sense of the categories a review might touch. Use that only as context for your own task list, not as a promise that any particular feature exists.

What to Record When a Step Has No Clear Outcome

The most valuable notes come from the moments where nothing obvious happens. When a step lacks a clear outcome, resist the urge to interpret it kindly. Write down what you saw, not what you assume was intended.

Use a short, repeatable note format. First, the task name. Second, what you did, in one sentence. Third, what appeared on screen, described literally. Fourth, what you expected instead. Fifth, a severity label chosen from a small fixed set, such as blocks the enquiry, confuses the reviewer, or cosmetic. Sixth, who owns the follow-up question.

Here is an example of the format in use, with invented content: task, find the sender's location; action, opened the enquiry record; observed, the location field was blank although the message mentioned a town; expected, the town shown or the field flagged as unread; severity, confuses the reviewer; owner, intake design.

Record the absence of a confirmation message, a missing status label, a button that does nothing visible, and any point where the reviewer had to ask the operator what to do next. That last one is often the clearest signal that the screen is carrying knowledge that only the builder has.

Do not record guesses about why something is missing, and do not record performance timings as findings unless you measured them in a consistent way. A single slow click on a shared laptop tells you very little. If speed matters to the review, plan a separate, simple measurement and say plainly that it is an informal observation.

Keeping the Enquiry Fictional and the Review Safe

The discipline of a fictional enquiry only works if everyone in the room treats it as fiction. Do not borrow a real message from the inbox, even with the names changed, because the details often survive the edit. Do not use a real phone number, a real address or a real email address in the worksheet. Placeholder values are fine and make the fiction obvious to anyone reading the notes later.

Keep credentials out of the review entirely. If the prototype needs a login, use a test account set up for the purpose, and do not write the password into the worksheet or the notes. If a task cannot be completed without access the team does not have, record that as a blocked task and move on.

Be careful with legal and regulatory questions. A review may surface questions about how enquiries are stored or how long they are kept, but the review itself is not legal advice, and the notes should say so. Route those questions to the appropriate person rather than settling them in the meeting.

Finally, agree before the session on what will happen to the notes. A short summary circulated within a few days, with the severity labels and owners intact, is usually more useful than a long transcript. Keep the fictional worksheet attached to the summary so the next reviewer can repeat the same walkthrough after changes are made.

Turning Review Notes into a Clear Brief

A completed worksheet and a set of notes give you something a build conversation can actually use: a described enquiry, a list of tasks, and a record of where the trace broke. That is more concrete than a general request to improve a system, and it keeps the discussion anchored to a real sequence of steps rather than a list of features.

Before approaching a software application development company, tidy the notes into three short lists: what worked, what blocked the enquiry, and what needs a decision from the business rather than from the developer. The third list matters most, because some gaps are design choices that only the business can settle, such as which enquiries should be routed to a person rather than answered automatically.

If the review is part of a broader look at how the website, enquiry handling and follow-up fit together, it can help to speak with a team that works across those areas. Blackstone Intelligence describes itself as building connected systems rather than isolated deliverables, and its services span AI automation, workflow automation, website development and software development. That breadth is useful context when you are deciding who to ask about the next step, but the review notes, not the service list, should drive the conversation.

Repeat the same fictional enquiry after any changes. Because the worksheet is fixed, a second walkthrough is comparable to the first, and the notes show plainly whether a previously blocked task now has a clear outcome. Keep the worksheet versioned and dated so the comparison stays honest.

A Short Checklist Before You Start

Use one invented enquiry with a clear expected outcome. For each field, write one task and an observable result. Assign an operator and a note-taker. Check that neither the sample nor the review notes contain real customer details or credentials. Date the worksheet and keep it for the next walkthrough.

Further reading

For the provider’s public service information, see Blackstone Intelligence’s software application development information.

Blackstone Intelligence is a technology consultancy based in Kuching, Sarawak, and operated by Blackstone Consultancy Sdn Bhd. It provides AI consulting and automation, web design and development, custom software development, and digital marketing services.