For agencies
You sell it. We salvage it.
Agencies are good at winning work, managing clients, and shipping the ordinary eighty percent. This is about the other twenty — the project that turned nasty, the codebase nobody wants to open, the incident on the Friday your senior is away.
The proposition
Your team handles the client. We handle the technical disasters.
A twelve-person agency can sell a seventy-five thousand dollar application and deliver it well. What it usually cannot do is carry a full-time principal engineer on the bench for the two months a year that go sideways — so when they do, the options are to burn the margin, burn the team, or tell the client the truth late.
Salvage is the third option: a senior escalation layer you can reach for by the week, by the project, or on a standing retainer. No recruitment, no bench cost, no long-term commitment, and no risk to the client relationship you spent years building.
When agencies call
-
The project that got away
Sold in good faith, six months in, and the architecture has stopped cooperating. Your team is capable and out of road, and the client is asking for dates nobody wants to give.
-
The senior who left
One person understood the whole system and now they work somewhere else. Everything still runs — right up until it doesn’t, and then nobody knows where to look.
-
The bid you’re not sure about
An RFP with a technical assumption buried in it. Twenty minutes of senior review before you sign is the cheapest insurance in this business.
-
The inherited client
You took on a retainer for a system somebody else built, and the handover was a repository URL and good luck.
-
The 4pm Friday
Production is misbehaving, your best developer is on holiday, and you need someone who has seen this shape of problem before.
The part that should matter most to you
We propose. You merge.
Salvage does not merge code into your projects. Every change arrives as a pull request with the reasoning attached, and one of your developers approves it.
The fear about subcontractors is usually described as "they'll take my client". That one is easy to write into a contract. The expensive one is quieter: six months on, there is a part of your own project that only the subcontractor understands, and every invoice after that reflects it.
This rule makes that outcome structurally impossible. Your engineer has to read and understand each change in order to ship it, which means the knowledge lands on your side of the relationship by construction rather than by promise. It is slower than handing us commit rights. It is also the difference between buying a fix and buying a dependency.
- Default
Read-only
Assessments, autopsies, code reviewNo write access is requested, because none is needed to tell you what is wrong. A surprising number of engagements never go past this tier — you get the diagnosis and your own team does the work.
- Standard
Propose, never merge
All repair and recovery workEvery change arrives as a branch, a pull request, and a written rationale for why it is the right change. Someone on your team reviews it and merges it. That review is not a formality — it is the mechanism that stops any part of your system ending up understood only by us.
- Break-glass
Time-boxed write
Outages, by prior arrangementRequires written authorisation from a named person on your side, runs for a fixed window, logs everything touched, and closes with a written handover of every change made. The log is part of what you receive, not a courtesy.
Worth saying plainly: this is a discipline, not timidity. There is a break-glass path for outages, agreed in advance and logged. What there isn't is a quiet accumulation of changes nobody on your team has read.
The rules of engagement
The thing you're actually worried about
Every agency that has ever subcontracted has the same fear: that the subcontractor ends up with the client. Here is the answer to that, stated plainly enough to hold you to.
You keep the client
Salvage does not contact your customer unless you ask, does not pitch them, and does not take them. Non-solicitation in writing on request, and gladly.
White-label by default
Work can land under your name, in your repository, in your project management tool, on your call. Plenty of engagements run with the end client never hearing the word Salvage.
We don’t sell what you sell
Salvage does not do greenfield builds, design, marketing sites, or retainers for ongoing feature work. There is no version of this where we become your competitor.
Your team gets better
Every change arrives as a pull request your developers review and merge, with the reasoning attached. They end up understanding the fix, which is how the next one of these becomes a job they can handle without us.
Three ways to work together
- On demand
One problem at a time
You call when something goes wrong. Scoped per incident, quoted before it starts. Nothing standing, nothing owed between jobs.
- Per project
Attached to one build
Salvage sits inside a single engagement for its duration — architecture up front, escalation throughout, and a hand on the difficult parts.
- Retained
Standing escalation
A monthly block of senior time across everything you have running. The cheapest of the three per hour, and the only one where problems get caught before they are problems.
Availability, stated honestly
Straight answers, not round the clock
Salvage does not sell 24/7 on-call, because a promise like that from a small shop is a promise that gets broken on the worst possible night. First replies come usually within a few business days. Dedicated cover is arranged in advance on active engagements, and honoured when it is.
Engagement capacity is deliberately small. If the calendar is full you will be told immediately rather than slow-walked — an agency that cannot get a straight answer about timing has been given no answer at all.
Become the agency that never gets stuck
No pitch deck and no procurement dance. Tell us what your team runs into, and we'll tell you whether Salvage is worth having in your contacts.
First reply usually within a few business days, from a person