Emergency Rescue Plan active — senior engineer replies within 12 hours See the plan →
Compare · people search: “freelancer vs agency for web design”, “who to hire for my website”

Freelancer, agency, or rescue — who should fix your website?

Three ways to hire help, three different risks. Here is how a business owner picks without learning to code.

Plain English · no jargon · 3-minute read

By Rodrigo Valenzuela Iturrieta, the engineer who does the work · Updated

Three ways to hire help, three different risk profiles — and none of them is objectively better. A freelancer is time on tap: cheapest per hour, great for a defined one-off task, and the riskiest when you can't see the meter or the finish line. An agency is a team on retainer: the steady-growth engine, and the slowest lane in an emergency. A flat-rate rescue is an outcome on a clock: one price, one engineer, 48 hours, and the guarantee that the fix is the fix. The honest answer depends on one question only — what kind of situation are you in right now?

A freelancer

$50–$150 / hour

time on tap — cheapest for a defined one-off task, riskiest in an emergency you can't scope

An agency (retainer)

$1,000–$5,000+ / month

the growth team for steady months — honestly the slow lane when something is on fire

A flat-rate rescue

$2,000–$25,000 · one price

the outcome on a clock — one senior engineer, locked price, 48 hours, proof in the handover

Eight questions that choose for you

  1. Is something broken right now? That overrides everything. A down, hacked, or crashing site is a rescue situation, not a hiring decision.

  2. Is it a defined one-off task? Add a page, tweak a form, fix a small thing — a freelancer is the fair, cheap answer. The risk only grows when the task can't be scoped.

  3. Do you want a number before you approve? A freelancer quotes hours, an agency quotes a relationship. A rescue quotes the outcome, in writing, up front.

  4. Can you afford the risk of the meter running? If a surprise $8,000 invoice would hurt, an open-ended hourly engagement is the wrong tool for an uncertain problem.

  5. Do you need proof it actually worked? Rescue handovers carry a log, before/after numbers, every credential still yours. A freelancer hands over time billed.

  6. Is the site healthy and improvement is the goal? Agency territory — genuinely. A retainer is for the good months, not the bad night.

  7. Is the problem one layer deep or the whole chain? Kernel to checkout — when the diagnosis crosses server, database, DNS, and code, you want one senior person who owns the whole chain, not a freelancer who knows one part.

  8. Worst case: does the person vanish? A freelancer can disappear mid-fix. A rescue's process is documented and rehearsed, so the work survives whoever's on the keyboard.

Which line describes you right now?

Something is broken — down, hacked, crashing, gone from Google?That's a fire → a rescue.
Yes — it's a fire. The 31 rescue tickets →
Healthy site, steady improvement is the goal?That's a retainer / agency job, honestly.
Start with the ideas in the guides → Or read the plain-English glossary →
A defined one-off task and you know the scope?Honestly, a freelancer is the fair answer.
Scope it first with the plain-English guides →

If the answer is a fire, send this tonight

Your words are enough: “checkout crashes,” “site was hacked,” “we vanished from Google,” “email isn't landing.” The reply is a flat number within 12 hours — the number is the number. That is asking a rescue what a freelancer can never answer and an agency won't answer: how much, by when, and what happens if it doesn't work.