When a contractor is the right answer
Contracting works well in three situations and badly outside them.
Bounded work with a real end. A migration, a website, a specific integration, a piece of design. Something you can describe as an output.
A skill you need occasionally. Nobody needs a full-time specialist for something that arises twice a year, and hiring one produces an underused person and a difficult conversation later.
Genuine uncertainty about whether a role is permanent. Using a contract engagement to find out is legitimate, provided you are honest about it with everyone including yourself.
Where it works badly is the open-ended role: "we need someone to handle marketing". Without a defined output, you get an expensive person waiting for direction, and the direction is the thing you did not have in the first place.
The test before you engage anyone
Can you describe what will exist at the end that does not exist now? If not, you have a role to define rather than work to contract, and no contractor will define it for you.
Scoping the work
This is where nearly all failures originate, and it is entirely within your control.
Describe the output, not the activity
"Twelve hours a week on marketing" buys you attendance. "A working email sequence that runs without manual steps, delivered by the 30th" buys you a thing.
Output-based scoping also makes disputes about value much less likely, because both sides agreed what existing meant.
The single most important stepWrite down what done means
Specific enough that you would both agree, on the day, whether it has happened. Include the quality bar, not just the existence of the thing.
If you cannot write it, that is the finding. Contracting a vague brief converts your ambiguity into someone else's invoice.
Checkable, not aspirationalSet review points before the end, not just at it
At least one checkpoint at roughly a third of the way through. Two on anything longer than a month.
A first review at the deadline means a mismatch is discovered when it is most expensive to fix, and everyone has already spent the budget.
A third of the way inSay what you will provide and by when
Access, information, decisions, sign-off, someone to answer questions. Contracts overrun far more often because the client was slow than because the contractor was.
Name who on your side owns this, with the same clarity you would want from them.
Your obligations, in writingAgree how change is handled
Scope changes on almost every engagement. Decide in advance whether that means a new quote, an hourly rate, or a documented trade against something else.
This conversation is easy before work starts and difficult in week three with a deadline approaching.
Before it happens
Choosing between candidates
Ask for work that resembles yours, not their best work. A portfolio shows their ceiling. What you need to know is what their ordinary work looks like on a project the size of yours.
Talk to a past client, briefly. The reference questions that work for employees work here too, and the most useful one is unchanged: what kind of client gets the best out of them? The full approach is in checking references properly.
Watch how they scope it back to you. A good contractor pushes on your brief, asks what you will provide, and names the risk they see. One who accepts a vague brief enthusiastically is either inexperienced or planning to bill the ambiguity.
Pay for a small paid trial where the engagement is significant. A week of real work tells you more than any interview, and it is fair to both sides.
Be honest about your own maturity. If you have no process and cannot make decisions quickly, say so. Some contractors work well in that environment and some need structure, and mismatches here are the second most common cause of a failed engagement after scoping.
Working together well
One point of contact on your side. Contractors receiving direction from three people produce work that satisfies none of them.
Answer questions fast. A blocked contractor is billing or waiting, and both are expensive. If you cannot respond within a day, say so up front so they can sequence around it.
Review at the checkpoints, properly. A skimmed review at week two produces a real review at week six, when changes cost several times as much.
Give feedback directly and early. The same structure that works for employees applies: specific, about the work, in the week it happened, as covered in giving feedback that doesn't damage trust.
Plan the handover from day one. What documentation, credentials, and context transfer to you at the end. Contract work that leaves knowledge only in the contractor's head recreates the single-person dependency you were probably trying to solve, and our guide to documenting a process covers what that handover needs.
The practical and legal points
These vary considerably by country, and the consequences generally fall on you rather than on the contractor. Get advice for your jurisdiction rather than relying on a summary.
Employment status. Broadly, the more you control how, when, and where someone works, whether they can send a substitute, and how integrated they are into your team, the more likely the relationship is to be treated as employment regardless of what the contract says. Getting this wrong can mean back taxes and employment rights you did not budget for.
Intellectual property. Do not assume you own what you paid for. In many jurisdictions IP in commissioned work stays with the creator unless assigned in writing. An explicit assignment clause is standard and worth having in your template.
Confidentiality and data. If they will touch customer data, the same obligations you carry apply, and that needs to be written down rather than assumed.
Payment terms and late payment. Agree them explicitly, including what happens if a milestone slips. Small suppliers are hurt badly by informal payment practices, and a reputation for slow payment makes good contractors decline you.
Insurance and liability. Check what they carry and what your contract says about liability caps, using the same framing as negotiating contract terms without a legal team.
One thing worth getting right once: have a standard contractor agreement drafted properly, with the IP assignment, confidentiality, and payment terms already in it. It turns every future engagement into a scoping conversation rather than a legal one.
Frequently asked questions
- When should you hire a contractor instead of an employee?
- When the work has a clear boundary and a defined end, when you need a skill you will not need continuously, or when you genuinely do not yet know whether the role is permanent. Contracting for an open-ended role usually costs more and produces less.
- How do you scope contract work properly?
- Define the output rather than the activity, name what done looks like, and set review points before the end. "Improve our onboarding" fails; "a working onboarding flow that a new customer can complete unaided, reviewed at weeks two and four" succeeds.
- Who owns the work a contractor produces?
- Whoever the contract says, and in the absence of a clause it may not be you. Intellectual property does not transfer automatically in many jurisdictions, so an explicit assignment clause matters and is worth getting right once.
- How do you avoid employment status problems?
- Take advice for your jurisdiction. Broadly, the more you control how, when and where someone works, the more likely the relationship is treated as employment regardless of what the contract says. The rules differ substantially by country and the consequences fall on you.
- Should contractors join your team meetings?
- Enough to do the work well, and not so much that they become a de facto employee. Beyond the practical point, heavy integration into your team structures is one of the factors that affects employment status assessments.