Selling Your Skills

Nobody recommends this one and the reason is structural rather than mysterious. You cannot sell a course about passive income on the back of “find a client and edit their videos”. There is no funnel in it, no screenshot of overnight revenue, nothing that works as a thumbnail.

So an entire genre of advice skips the only route that reliably pays inside a fortnight, and points you instead at the ones that pay somewhere around month nine.

You can be paid within two weeks of deciding to try this. No audience, no algorithm, no waiting for a channel to mature. There is no other route on this site where that is true.

This page is long because the useful part is the specifics: where the clients actually are, what to send them, what to charge, and what happens when they reply.

Who is buying, and what they need#

The demand comes from a group that grew very quickly and has no production capacity: small creators and small businesses who suddenly have money.

Creator advertising spend in the US reached about $44 billion in 2026, up from $37 billion the year before, and nano and micro creators now take 49.9% of it, up from under a fifth a few years ago. That last number is the one that matters to you. It means tens of thousands of people with a few thousand followers each are now being paid by brands, on deadlines, with no team.

A creator in that position has a specific problem. They can make the thing but they cannot make it fast enough, and the brand wants three versions by Thursday. What they buy is throughput and reliability, not artistry.

The work they buy, roughly in order of how often it comes up:

• Cutting long footage into short clips. The single most common paid job in this market. A podcast or long video goes in, eight to twelve vertical clips come out, captioned and hooked. It is repetitive, it is learnable in a fortnight, and it is in constant demand because every creator with long-form content knows they should be doing it and almost none of them want to.
• Editing long-form video. Higher skill, higher rate, slower to win. Clients are more careful about who touches their main channel.
• Thumbnails. Small individually, but it turns into a subscription-shaped arrangement quickly, because a channel needs one or two a week forever.
• Running social accounts. Scheduling, posting, replying, light design. Less about craft than about someone else not having to remember.
• Podcast production. Audio cleanup, chapters, show notes, distribution. A narrow field with less competition than video.
• Newsletter writing and editing. Undersupplied, because most people who can write well would rather write their own thing.
• Research and scripting. The hardest to sell cold and the best paid once trusted, since it requires the client to believe you understand their audience.

If you are choosing where to start with no experience, start with clipping. It has the shortest learning curve, the most demand, and it puts you inside a client’s operation where the other work becomes visible.

What it pays#

These are 2026 market ranges. Read the reliability note under the table before you use them, because the freelance market has no central data source and every published figure is somebody’s compilation.

Rate guides, not measured transactions. Sources: goLance on editor rates, goLance on social media rates, SolidGigs, Content Beta, VisualKit. Rate progressions, the retainer sheet, the scope account and the agency fee demand are self-reported by working freelancers on public forums, not measured data: podcast production rates, the Melbourne retainer sheet and scope account, the agency service fee. Australian dollars where stated.

What that looks like as a monthly income#

The ranges above are per unit, which is not how you will think about it after the second month. Here is the arithmetic that matters.

Suppose you clip long videos into shorts. You charge $150 for a batch of ten clips from one source video. Early on a batch takes you six hours, so you are earning $25 an hour, the bottom of the range, which is where everyone starts. Two clients sending two videos a month each is four batches, $600 a month, 24 hours of work.

By month four the same batch takes you two and a half hours, because you have templates for captions, a repeatable structure for finding the hook, and you have stopped rewatching footage you have already seen. Same $150, now $60 an hour. This is the moment that decides whether this route works for you, and it turns on one thing: whether you charge per batch or per hour. Per batch, your effective rate has more than doubled. Per hour, you have given the entire gain to the client and you are earning $150 for the day instead of $360.

Take it one step further. Four clients on a monthly retainer at $500 each, four videos a month per client, is $2,000 a month for roughly 40 hours of work. That is not a large income, and it is a stable one arriving on the first of the month, which is a different thing from the same figure arriving in unpredictable lumps. It is also enough runway to build something of your own, which for most people reading this is the actual point.

Where the clients are, specifically#

This is where most advice becomes useless, so let us be concrete. There are four places, and they are not equally good.

Approaching creators directly#

The people who need this work are visible and their contact details are public. Open YouTube, search a niche you understand, and filter for channels between roughly 5,000 and 100,000 subscribers who upload long-form video and post shorts irregularly or not at all. That gap, long videos, no shorts, is a channel with unused material and someone who knows it.

Their business email is usually in the channel’s About tab under “View email address”. If it is not, they often have a website, a Linktree, or a newsletter with a reply address that reaches them.

Twenty to forty of these, approached properly, is a realistic first client. The response rate is low and the quality of response is high, because you are talking to the decision-maker with no intermediary and no bidding.

Marketplaces: they work, at a price#

Upwork, Fiverr and the rest do supply real work, and the trade is visible in the numbers: Fiverr averages around $18 to $19 for a thumbnail job, well below what the same work earns from a direct client. You are competing on price against a global market, with the platform taking a cut.

They are useful for one specific thing: producing your first two or three paid jobs and the reviews that go with them, when you have no portfolio and no credibility. Use them for that and then leave. People who stay tend to stay at the bottom of the range for years.

Communities where creators talk to each other#

Discord servers, subreddits and Slack groups for particular niches, podcasters, newsletter writers, a specific game, a specific software, frequently have a hiring channel, and the people posting in it are pre-qualified: they have decided to pay someone and are looking for who.

The thing that works here is not posting that you are available. It is answering technical questions for a few weeks until people recognise your name, at which point work arrives without you asking. It is slower than cold email and the clients are better, because they came to you.

Agencies and other freelancers, the route nobody mentions#

Small agencies and busy freelancers subcontract constantly and quietly. An editor with more work than hours would rather hand you the overflow at 60% of their rate than turn a client down. The pay is lower and the work is continuous, and there is no pitching involved once you are in.

Approach freelancers slightly above your own level, not agencies. A one-line message saying you handle overflow, with two samples, is enough. Most will say no and remember you in three months when a deadline collapses.

The portfolio problem, and how to solve it in a weekend#

You need three to five samples and you have no clients. Everyone starts here, and the solution is that nobody asks whether the client was real. They look at whether the work is good.

Make the samples yourself, from public material:

• Clipping: take a long video from a channel you would like to work with. Cut three shorts from it. Publish them nowhere; put them in a folder.
• Editing: take four minutes of someone’s raw or loosely edited footage and re-cut it. Show the before and after side by side, with a sentence on each decision. The sentence matters more than the edit, because it demonstrates that you make choices rather than following defaults.
• Thumbnails: redesign three from one channel. Keep the original next to yours.
• Social management: take a real account and write a two-week posting plan with the actual posts drafted, not described.

Two rules. Use real channels rather than invented briefs, because it proves you can work with the material that actually exists. And target the samples at the client you are about to approach, since a sample of their own content is a far stronger pitch than a general showreel.

The email, in full#

Most outreach fails for one reason: it is about the sender. Here is a version that does not do that, which you should adapt rather than copy word for word.

Subject: three shorts from your Kubernetes episode

Hi Sam,

I watched the Kubernetes migration episode last week. The section about the failed rollback at 14:20 is the best explanation of that problem I have heard, and it is buried three quarters of the way into a fifty-minute video.

I cut it into three vertical clips, hook, problem, resolution, with captions. No charge and nothing expected: [link]. Use them if they are useful.

I do this for a handful of technical channels. Ten clips from one source video, captioned and ready to post, is $150 with a three-day turnaround. If that is interesting I can do your next episode.

Either way, the rollback section deserves an audience.

Alex

The blank version. The bracketed parts are the only ones that change, and the second line is the one that decides whether you get a reply:

Subject: [what you made] from your [their specific episode]

Hi [name],

I watched [the specific episode] last week. [One sentence about
a specific moment, with a timestamp.]

I [did the thing] and it's here: [link]. No charge and nothing
expected. Use it if it's useful.

I do this for [the kind of client they are]. [Deliverable] is
$[price] with a [n]-day turnaround. If that's interesting I can
do your next [episode/week/batch].

Either way, [the specific moment] deserves an audience.

[your name]

One thing to hold on to while adapting it: the timestamp is not decoration. It is the only line in the email that cannot be produced at scale, which is exactly why it is the line that earns a reply.

Why this works, since the mechanics are the transferable part:

The subject line is about their content, not your services. It survives a glance at an inbox.

The first paragraph proves you watched the video, with a timestamp. This is the part that cannot be faked at scale and is therefore the part that gets replies. Three sentences of specific observation outperform any amount of describing yourself.

The work is already done. You are not asking them to imagine what you might produce, or to book a call, or to look at a showreel of other people’s channels. The thing exists and they can watch it in forty seconds.

The price is stated. Most freelance outreach hides the price to open a conversation, which mainly filters for people who enjoy conversations. A price in the first email means the reply you get is from someone who has already decided it is affordable.

And it closes without pressure. No follow-up sequence, no “just checking in” three days later. One good email, and if nothing comes back, move on.

Expect roughly one reply for every eight to ten emails of this quality, and perhaps a third of those replies to become work. Which is why the number is twenty to forty, and why sending a hundred generic versions of the same message performs worse than twenty specific ones. The specificity is the entire product.

What happens when someone says yes#

Before you start, three things go in writing. An email is writing; it does not need to be a contract.

What you are delivering. Not “shorts” but “ten vertical clips, 30 to 60 seconds, burned-in captions, delivered as MP4 in a shared folder”.

Two rounds of revisions, then an hourly rate. This single sentence is the difference between a profitable job and one that never ends. Say what a round is: one set of notes, returned together, not a message every hour for a week.

When you get paid. Half up front from anyone you have not worked with, the balance on delivery. You will feel awkward asking a stranger for a deposit. Ask anyway. A client who will not pay $75 up front to a new supplier is a client who will not pay $150 afterwards, and finding that out now costs you nothing.

Then deliver early. Not on time, early. The first job is not really about the edit; it is about whether working with you is uneventful. Turn it in a day ahead, in the format they asked for, in the folder they specified, with a one-line note saying what you would do differently next time. Reliability is the product being purchased and it is in short supply.

Turning one job into steady income#

Single jobs are exhausting because every one requires a new sale. The transition to retainer is the whole game, and it happens in a specific conversation after roughly the third delivery.

What you say, in substance: “This is the third batch. Rather than invoicing each one, I could take four videos a month at $500 and you would stop having to ask. Same turnaround, first refusal on your calendar.”

Note what is being sold. Not a discount, the per-unit price is similar or slightly better for you. What the client is buying is not thinking about it, and reserved capacity. What you are buying is predictable revenue and no repeat sale, which is worth more to you than the extra $50.

One published account of where that ends up. A podcast producer who started at “50$ per episode” now charges “$650-$1000 to produce two episodes in a month”, which includes travelling up to an hour, setting up, breaking down and being the only technician on site. Same work, restructured from per-episode to per-month, and the clients came inbound from having run his own show rather than from outreach.

A videographer in Melbourne publishes his retainer sheet, which is useful because it shows what a retainer actually contains at three sizes. In Australian dollars, per month:

Self-reported, August 2026. What matters more than the numbers is the line underneath them: he excludes “account management, community management, Canva graphics, ads” in writing. Every one of those is something a client will otherwise assume is included.

He also shows the cost of discounting while new. He proposed $1,500 a month and “because I was new to offering this as a service, I let them negotiate me down to $750/month”. He is still at $750 per clinic, and the increase he is now planning is back to roughly his original ask. A first-client discount is not a discount on one job; it is the ceiling you then spend a long time climbing back to.

Two or three retainers make this a job with an income. Five make it a business you may not have intended to start, which brings its own decision, covered further down.

Raising your rate without losing anyone#

Almost everybody sets a first price by guessing, then keeps it for two years.

Raise on new clients only. It is easier, it is not a confrontation, and it lifts your average steadily without risking the work you have. Quote the new number to the next person who asks, and if they say yes without hesitating, the number was too low and you should raise it again for the person after that. Keep going until someone declines. If nobody has ever declined your rate, it is below the market and you are subsidising your clients.

Existing clients get raised at a natural boundary, a new year, a renewal, a change in scope, with one sentence of notice and no apology or justification. “From January the retainer is $600.” Most say fine. The one who leaves was your worst client, and the hour they free up is worth more than they paid you.

One thing not to do: never raise a rate mid-project, and never renegotiate a price you have already agreed. Deliver what you promised at what you promised, and price the next one properly.

The clients to refuse, and how you spot them#

These come up often enough to be worth memorising, because the cost is paid in weeks rather than money.

“We’re looking for someone long-term, so the first one should be free as a test.” A paid trial is normal and reasonable. An unpaid one from someone who intends to pay later is a filter for people who work for free, and the long-term relationship does not appear.

“Can you send a sample edit of our footage before we decide?” Fine once, briefly, if you chose to make it as outreach. Not fine as an unpaid audition against four other freelancers doing the same job. Offer a short paid pilot instead.

“We’ll pay in exposure. I’ll credit you to my audience.” A credit is worth having and it is not money. If someone is monetising the content you are working on, they can pay for the work.

An agency that asks you to pay a fee. One editor delivered work through an agency middleman, waited a week, chased twice, and was then told “i need to pay them a ‘service fee’ of $15 before they send my contact details to the client”. They offered screenshots as proof the client existed. It is an advance-fee scam pointed at the freelancer instead of the customer, and the amount is small on purpose, because $15 feels cheaper than arguing. Money moves from client to you through an agency, never the other way. The tell he noticed: mid-conversation they “suddenly replied in italian”.

“Just make it pop.” Not malice, but a client who cannot describe what they want will not recognise it when it arrives, and you will discover this during revision round four. Push for a reference: a video they wish theirs looked like. If they cannot produce one, price the ambiguity in or decline.

“Our payment terms are net 60.” Common from anything agency-shaped, and it means you are financing a company larger than you for two months. Negotiate to net 30, take a deposit, or decline. Late payment is normal in this market and being relaxed about it is expensive.

Anyone who negotiates hard on the first job. Not because negotiating is wrong, but because in this market the client who fights hardest over $150 is reliably the one who sends notes at midnight and pays in six weeks. The correlation is unreasonable and it holds.

Where the money leaks#

Hourly pricing you never leave. The mechanism is in the arithmetic above: hourly billing pays you less as you get better, which is the wrong incentive to build a career on. Move to per-project as soon as you can estimate a job within about 20%, which for most people is after five or six of them.

Scope with no edges. “Ten clips” becomes “and the thumbnails” becomes “and could you schedule them”. Each addition is small and reasonable and none of them are paid. Say yes to one of them, once, as goodwill, and then say “happy to, that is an extra $40” every time after. Nobody is offended by this. They ask because most freelancers do it for free.

The clearest published account of how that happens is worth reading in the person’s own words, because nobody thinks it will happen to them: “There hasn’t really been a properly defined scope since we started, and looking back, I really wish I had established clearer boundaries from the beginning. Because I was new and wanted to do a good job, I found myself saying yes to a lot of things outside the core filming/editing work. Over time, it created an expectation that I would basically handle whatever was needed on the content/social side.”

Note the mechanism. Not one large concession but a series of small yeses, made while trying to be good at the job, which harden into an expectation. The same person absorbs unpaid reshoots because the client books him into lunch breaks and then “staff/rooms aren’t available”, which is scope creep arriving through the diary rather than the brief. Both are fixed by the same thing: a written list of what the fee covers, sent before the first invoice rather than after the third.

One client at more than half your income. This is the same concentration risk as a creator dependent on one platform, with fewer options, and it ends abruptly when that client’s own income changes. Three clients minimum, and treat the arrival of a fourth as more valuable than a rate rise.

Working for free while calling it marketing. One targeted sample as outreach is marketing. Three finished edits for someone who might hire you is unpaid labour with a story attached.

What AI changed here#

AI compressed the bottom of this market and left the top of it alone. Both halves matter to you.

The compression is real. Video teams shrank from eight or ten technical roles to three to five strategic ones, cutting overhead by around 41%. Basic cutting, captioning, transcription, rough assembly, translation and silence removal are now close to free and available in a browser. Nobody will pay you a competitive rate for something a tool does in ninety seconds, and there is no argument to be made about craft that changes this.

What held its value, and in some cases gained: the decisions. Which forty seconds of a fifty-minute video is the one worth cutting. Where the first frame should land. Whether the hook is doing anything. What to cut entirely. These are judgements about a specific audience, and the tools do not have an opinion about your client’s audience.

The practical consequence is a change to how you describe what you sell. “I will edit your video” is a description of a task that software performs. “I will find the three moments in your episode worth clipping and cut them so people stay past the first second” is a description of a judgement, and it is the same job, priced differently, won differently, and considerably harder to replace.

Use the tools. Use them heavily, they are why a batch takes two hours instead of six, and that gain belongs to you as long as you are not billing hourly. Charge for the part that requires having watched the whole thing and having a view about it.

One more effect, less discussed: your clients are now subject to platform rules that punish output which looks mass-produced. YouTube’s Inauthentic Content policy since July 2026 excludes templated, mass-produced material from monetisation by name. If you deliver work that trips that, you have cost your client their income, which ends the relationship in the worst possible way. Read what gets you demonetised before you build a workflow around volume, because it is now part of your job to know.

Who should not do this#

Anyone hoping for income that continues when they stop working. This is a job with better hours and no commute. It does not compound, there is no asset at the end of it, and it never becomes passive. Fifty completed jobs leave you with a portfolio and no audience, archive or product.

Anyone who struggles with deadlines. One missed delivery costs more than the fee, and niches are small enough that it travels.

Anyone whose real ambition is their own audience, with no spare hours. This is the trap that catches most people, and it is worth being honest about: client work absorbs precisely the hours you would otherwise spend building your own thing, it pays immediately, and the pull towards taking one more job is constant. Plenty of people intend to freelance for six months to fund a project and are still doing it four years later.

The version that works is deciding the boundary in advance, this many hours a week, this many clients, no more and treating a request to exceed it as a request to postpone your own work by a month. And it is.

The first fortnight, concretely#

1. Day one. Pick one service. Not video and design and social. Clipping if you have no strong preference.
2. Days two and three. Make three samples from public material on channels you would like to work with.
3. Day four. One page: what you do, three samples, a price, an email address. No logo, no brand, no about-me paragraph.
4. Days five to eight. Build a list of thirty channels in one niche, long-form video, few or no shorts, 5,000 to 100,000 subscribers, with a business email for each.
5. Days nine to twelve. Send twenty emails on the pattern above. Watch each video before writing. Make the free clip for the ten you most want to work with.
6. Day thirteen onwards. Answer replies within the hour. Agree deliverables, two revision rounds and half up front in an email. Deliver a day early.
7. After the third delivery with any client. Propose the retainer.
8. After the first client. Raise the price for the second. Repeat until someone says no.
9. At three clients. Fix how many hours a week this gets, and defend the remainder.

The rate calculator will price a specific job including revisions and rush turnaround. Before you agree to anything with a company rather than an individual, contract clauses that cost you money is ten minutes well spent, since scope and revision terms cause more losses here than in any other route. And once there is money coming in, money basics covers what to do with irregular income before it disappears.