Brendan ElgieSoftware & Systems Consultant

02 · Approach

Most of the risk in custom software is structural.

Late, over budget, abandoned when the one person who understood it left. Those aren’t coding problems. They’re problems with how the work was set up, so that’s where I deal with them.

03

The phases

  1. 01

    Discovery

    One to two weeks · fixed fee

    I sit with the people doing the work and follow the job end to end: the quote, the schedule, the invoice, the spreadsheet nobody admits to. What comes out is a written map of where time and information actually leak, costed and ranked.

    You getA document you own, whether or not you hire me to build anything.

  2. 02

    Shaping

    Days, not weeks

    We pick the smallest thing that would make a real difference and define it precisely. What it does, what it deliberately doesn't do, what it costs, and when it lands.

    You getA fixed scope and a fixed price, agreed before anyone writes code.

  3. 03

    Build

    Typically four to ten weeks

    Weekly working software, in your hands, on real data as early as it's safe to. You'll be bored of seeing it before it launches, which is the point. Surprises at the end are the expensive kind.

    You getSomething in production, being used, not a demo.

  4. 04

    Handover

    Built in, not bolted on

    Documentation written for whoever comes next, a walkthrough session, and every account, domain and repository transferred into your name.

    You getYou could replace me tomorrow without drama. That is the standard.

04

Principles

A few things I won’t bend on.

Smallest thing that works

Most failed software projects were too big on day one. I would rather ship something narrow that earns its keep in a month than something comprehensive that arrives in a year and fits nobody.

Boring technology, on purpose

I pick tools that will still be maintainable in five years by someone who isn't me. That usually means fewer moving parts, static where possible, and no framework chosen because it was interesting.

Your data stays yours

You own the accounts, the repository, the domain and the database. Canadian data residency where the work calls for it. Nothing about the arrangement should make leaving difficult.

Automate the repeatable, not the judgement

Software should take the retyping, the chasing and the reconciling. The decisions that need someone who knows the business should stay with the person who knows the business.

If it's not worth building, I'll say so

Sometimes the answer is a spreadsheet used properly, or a setting in software you already pay for. Telling you that costs me a project and earns the next three.

05

Practical questions

CostDiscovery is a fixed fee. Build work is quoted as a fixed price against a fixed scope, invoiced in stages. I don’t bill hourly, because it puts us on opposite sides of every decision.
SizeMost engagements land somewhere between a few thousand and the low tens of thousands. If your problem is smaller than that, I’ll tell you and probably point you at something off the shelf.
TimelineDiscovery starts within a couple of weeks, usually. Build timelines depend on scope, but I’d rather quote eight weeks and mean it than quote four and not.
After launchOptional support on a monthly basis, or nothing at all. The handover is written so walking away is genuinely an option for you.
LocationBased in Sault Ste. Marie. In person across the Algoma District, remote everywhere else in Canada. For operational work, being able to stand in the room matters more than people expect.

Next step

Start with the discovery.

It’s the cheapest way to find out whether there’s a project here at all, and you keep the findings either way.

Start a conversation bren.elgie@gmail.com