Scoped options

Three doors instead of one number

A fund's back-office module needed rebuilding and the budget was not stated. I priced three genuinely different scopes rather than guessing which one they meant.

Low-cost, balanced and complete — differing in content, not in padding
3 optionsLow-cost, balanced and complete — differing in content, not in padding
The honest spread between the smallest and largest scope
240 → 400 hoursThe honest spread between the smallest and largest scope
Regression coverage written down for the module next door
Checklist, not memoryRegression coverage written down for the module next door

The client

A European venture fund's internal investor platform — onboarding and know-your-customer checks, capital calls, distributions, statements and investor communications — maintained as a stream of specified improvements rather than as a single project.

The engagement

A flow of analyst-led improvements on time and materials, with each significant module priced as a set of options before work begins.

The problem

When a client will not state a budget, most vendors guess, and both guesses are wrong: too high loses the work, too low wins work that will be argued about later. Meanwhile a back-office module in a fund platform touches permissions, audit history and bank details, so 'small' and 'complete' are genuinely different products rather than the same product at different quality.

What I did

I priced three doors. Each option is a real scope with its own content — what statuses exist, whether changes are logged to the activity trail, whether rejection carries a reason and a filter, whether permissions cover it, whether history and callbacks are included — so the client is choosing capability rather than negotiating a discount. Because the platform is maintained as a stream rather than a project, I also wrote a regression checklist for the adjacent communications module, covering the things that break silently: template name collisions, which data points may be used in a subject line, personalisation files with empty cells that must block a send rather than send a blank. And the specifications were written against the client technical lead's own review comments — bitmask permissions, a single system-wide activity log, invitation tokens rather than passwords — because a specification that ignores the reviewer's points gets rewritten anyway.

What was built

Three costed scopes for the same module, differing in what they actually include rather than in padding; a documented regression checklist for the neighbouring module so quality did not depend on memory; and requirements written to a technical lead's review comments rather than around them.

On the table at the end

  • Three costed scope options with hours, price and content per option
  • Regression testing checklist for the adjacent module
  • Specifications reflecting the client technical lead's review points
  • Weekly time-tracking detail by activity

What it changed

Let a client with an unstated budget choose a scope rather than negotiate a number — a low-cost option, a balanced one and a complete one, each with its own hours, price and explicit content, so the trade-off was visible instead of implied.

How it ran

  1. 01

    Price capability, not padding

    Three scopes that genuinely differ in what the module does, so the client compares outcomes rather than discounts.

  2. 02

    Make the audit trail explicit

    Whether status changes are logged, whether rejection carries a reason — stated per option because in a fund platform this is the difference.

  3. 03

    Write the regression down

    A checklist for the adjacent module covering the silent failures, especially personalisation with missing values.

  4. 04

    Answer the reviewer

    Specifications written to the client technical lead's review points rather than around them.

  5. 05

    Track time by activity

    Weekly detail by task on a time-and-materials stream, so the invoice is readable without a meeting.

Something similar on your plate?

Thirty minutes, no deck. I will tell you whether it is worth doing at all.