Approach
How a Wisqora project actually runs.
No discovery phase that bills for three months and produces a slide deck. The first step costs you nothing and ends with a document you keep. The second puts something real into production inside two weeks.
The engagement
Four steps, and what each one owes you.
Map
Week zero · no charge
One call, then a short written document: where your team’s hours are actually going, which of those hours a machine should be doing, and what each one is costing you a month.
You keep that document whether or not you hire us. If the honest answer is that your biggest problem isn’t an engineering problem, it will say so.
You come out with
- A ranked list of manual work, with a rough cost against each line
- A clear recommendation for what to do first
- An estimate - range, not a single number, because single numbers at this stage are guesses wearing a suit
Pilot
Weeks 1-2 · fixed scope
We take the single highest-cost item off that list and automate it. Not a prototype, not a demo - the real thing, in production, with the before-and-after measured.
Two weeks is deliberate. It’s long enough to deliver something that genuinely runs and short enough that you find out cheaply whether working together is a good idea.
You come out with
- One piece of manual work, gone, running in your accounts
- A measured before and after, so the value isn’t a matter of opinion
- An informed decision about whether there’s a bigger project here
Ship
Weeks 3-8 · scoped from the pilot
Now the rest gets built, around something that already proved itself. Scope comes from what the pilot taught us, not from what the first conversation assumed.
You see working software every week. Nothing is held back for a big reveal at the end - big reveals are where projects go to fail quietly.
You come out with
- The remaining automations, agents, product or platform work
- Weekly working increments in an environment you can open yourself
- Monitoring and alerting from day one, not added afterwards
Hand over
Final week · and afterwards
Everything gets handed across: code in your repository, infrastructure in your accounts, credentials in your password manager, and a runbook written for whoever picks it up next - which may well not be us.
An optional support retainer exists if you want one. It is not a condition of anything working.
You come out with
- Documentation written for the next engineer, not for the invoice
- A runbook covering the failure modes most likely to matter
- No lock-in: nothing here requires Wisqora to keep running
Principles
Six rules we don’t break.
These exist because fourteen years inside enterprise delivery is fourteen years of watching what happens when they’re ignored.
We measure before we build
If we can’t say what something costs you today, we can’t honestly claim to have improved it. Measurement comes first, every time.
One thing in production beats ten in a deck
A pilot that genuinely runs tells you more about a working relationship than any proposal document ever will.
You own all of it
Code in your repository, infrastructure in your accounts, documentation written for whoever comes next. Nothing depends on us staying.
Boring technology where it counts
Novelty belongs in the parts of the system where being wrong is cheap. The layer everything else sits on should be dull and well understood.
We’ll tell you it’s a two-day fix
Sometimes the honest answer is small. Saying so costs us an invoice and earns the next three.
Small number of projects at a time
Wisqora runs lean by design, so you always talk to whoever is doing the work - and capacity is genuinely limited. We take on a small number of engagements at once to keep every one senior from first call to handover.
Boundaries
Three things we say no to.
A short list, but it saves everyone a month.
Hourly bodies on a team
We don’t do staff augmentation. We take responsibility for an outcome, or we don’t take the work.
Rip-and-replace of something that works
The most expensive projects usually start by throwing away a system that was doing its job badly but reliably.
AI where a rule would do
If a lookup table and an if-statement solve it, that’s what you should have. It’s cheaper, faster and it never hallucinates.
The point of hiring an engineer is to end up with fewer things to think about, not more.
Step one
The map costs you nothing.
One call, then a written breakdown of where your hours are going and what they cost. You keep it whether or not anything else follows.