Most outsourcing decisions get made on price alone, and that is usually where they go wrong. A cheaper delivery route that removes your only view of a process, or hands a customer-facing task to someone without context, can cost more than it saves within a quarter.

The framework below works through four stages in order. Answer the questions honestly before you look at any quote. If a task fails the first stage, no price fixes it.

Stage 1: Is this work replaceable?

The first question is not “what does it cost” but “what happens if someone other than you does it”. Separate your work into three groups.

Core, differentiating work

Activity that encodes your judgment, relationships or proprietary method — the thing a customer is actually paying for. Keep this. Outsourcing it transfers your competitive position to a supplier along with the task.

Supporting work that still needs your context

Billing, client onboarding, anything where a mistake is visible to a customer or affects compliance. This can be outsourced, but only with documented rules, a named reviewer on your side and a defined quality check. If you cannot write the rule down, you are not ready to hand it over.

Commodity and volume work

Repeatable, high-volume, rule-based tasks: data entry, catalogue updates, list building, inbox triage, report assembly. These are the strongest candidates, because the value of keeping them in-house is mostly habit rather than control.

Stage 2: Can you describe it clearly enough?

Outsourcing succeeds when the work has a definition of done. Before you commit, write a one-page brief covering six things:

  1. Input — what you provide, in what format, how often.
  2. Output — exactly what is delivered back, in what format.
  3. Rule set — the decisions that are fixed, and the ones that need judgement.
  4. Exceptions — the situations to escalate rather than resolve.
  5. Volume and rhythm — typical quantity, and what happens in a peak.
  6. Acceptance — how you will check the work, and what a rework looks like.

If you cannot produce that page, the task is not underspecified in the market — it is underspecified in your business. Documenting it is often worth more than outsourcing it, because it usually reveals that some of the work should simply stop.

Stage 3: Compare the real cost, not the headline price

In-house cost is rarely just a salary. For a fair comparison, build up both sides on the same basis for the same period.

Cost elementIn-houseOutsourced
Direct labourSalary, employer costs, benefitsAgreed fee for scope or time
Management timeYour hours supervising, reviewing, correctingYour hours briefing and accepting
Cover and resilienceSickness, leave, vacancy periods, ramp-upSupplier continuity, substitution terms
ToolingLicences, hardware, seat costsOften supplier-provided; check what you must still pay for
Quality failuresRework you absorb silentlyRework under a defined acceptance rule
ExitRedundancy cost, knowledge lossNotice period, handover, document return
Compare on the same period and the same volume, or the numbers will not mean anything.

The management-time row is the one people skip. If outsourcing a task leaves you spending as long briefing and checking as you did doing it, you have moved the work, not removed it. That can still be the right choice — but it should be an explicit one, made for quality, cover or speed rather than cost.

Stage 4: Test before you commit

Never start an outsourcing relationship at full scope. Run a bounded first piece of work — a sample batch, a single report cycle, one week of a recurring task — with the same brief, the same acceptance rule and the same escalation route you intend to use permanently.

A trial answers three questions that no proposal can: whether your documentation is actually clear enough, whether the supplier follows it under real volumes, and whether the review rhythm fits your week. Where a supplier will not agree to a limited first piece of work before you commit, treat that as an answer.

Where hiring in-house is still the better answer

Outsourcing is not the default. Hire in-house when the work needs daily judgement you cannot write down, when it is tightly coupled to other work only you do, when it is small and irregular enough that briefing would cost more than doing, or when you are still discovering what the task should be. Outsourcing a process you have not yet designed usually just adds a supplier to the confusion.

A short decision summary

  • Outsource confidently when the work is repeatable, rule-based and describable.
  • Outsource carefully when the work is supporting but customer-visible — with documented rules and your own reviewer retained.
  • Keep in-house what carries your judgement, relationships or method.
  • Compare total cost over the same period, including your own management time.
  • Trial a bounded piece before committing to recurring scope.

Next step

If you have a specific task in mind and want to know whether it is a sensible candidate, send us a short description of the input, the output and roughly how often it occurs. We will tell you whether we can take it on, and if not, why. There is more on how we work, on the pricing page, and in the six services we run often enough to have a documented standard for each.