How to start UGC: a practical first month

Days one to seven, ending with one finished spec piece

Choose your category by what you can film happening, not by what you like. A product that visibly changes state in under ten seconds gives you something to point a camera at; a product whose benefit is invisible gives you a monologue. Foam, dissolve, steep, stretch, lift, seal, fit: those verbs are the shortlist.

Three examples of the kind of thing that works from a standing start. A shampoo bar lathering in wet hands, where the whole point is whether it actually foams. Laundry detergent sheets going into a drum, because the sheet disappearing is the demonstration. A cold brew concentrate poured into milk, where the colour bloom does the work for you. Pick whichever category holds three products you already own.

1. Day one, inventoryWalk your flat and write down six products that visibly do something. Whichever category claims three of them is your category for the next month. Stop deliberating at that point.
2. Day two, twenty five wordsWrite an opening line, three beats and a closing line. Not a script, twenty five words total. A script generator is useful here only for the shape; rewrite every line into how you actually speak.
3. Day three, shoot four shotsWide on you and the product, tight on hands, tight on the change happening, and one final held frame of the label. Take each one eight or ten times. Expect the first six takes to be unusable.
4. Day four, cut it shortAssemble under thirty seconds in whatever editor is on your phone, burn in captions, and export without a watermark from any app that would add one.
5. Day five, watch it wrongPlay it on a phone, standing up, sound off, then again with sound and no picture. Write down the exact second you stopped caring. There is always one.
6. Day six, refilm one beatOnly the beat that failed. Beginners rebuild the whole video and lose a week; the fix is almost always the opening two seconds or the shot where the change happens.
7. Day seven, post and filePost it, pin it, and save the originals in a dated folder. Write one sentence describing what the piece demonstrates, because you will need that sentence when you start reaching out.

Twenty rows is the smallest sample that says anything. Ten tells you nothing at all.

Gear, decided by what a viewer can actually detect

The equipment argument dissolves once you ask what survives into the file. Nobody watching can see your tripod, your light panel or your lens attachment. What they can see and hear is a short list, and every item on it can be fixed for almost nothing.

Spend money in this order when you have it: something that holds the phone still, a clip-on microphone if you talk in a room with hard floors and no curtains, and a sheet of white card to bounce window light back into your face. Do not buy a gimbal, a camera, a lens kit, a preset pack or an editing subscription in month one. None of them fixes anything on the list below, and all of them delay day three.

• Which direction the light came fromWindow light on your face reads as daylight. Ceiling light reads as a rented room. Film between mid morning and mid afternoon, facing the window, and the single biggest quality gap closes for free.
• Whether the room echoesHard floors and bare walls put a slap on your voice that no amount of editing removes. Film in a room with a rug, a sofa or hanging clothes in it, or hold the phone closer and speak more quietly.
• Whether the frame wobblesHandheld is fine and expected, drifting is not. Brace your elbow against your ribs or lean the phone against something. A viewer forgives a shake and switches off during a slow float.
• Whether the label is readableA buyer needs one frame where the product name is legible at thumbnail size. Rotate the bottle until the text faces the lens, even when that feels staged, and hold the shot for a full second.
• What is behind youTwo minutes clearing a worktop beats any lighting purchase. Leave enough real mess that it looks like a home and not enough that the eye goes to the drying rack instead of your hands.

Three pieces, then six, then widen

One spec video proves you can make a video. It does not prove you can be relied on to make another one next Tuesday, which is the actual question a buyer is asking. Three pieces in the same category, in three different shapes, answers it. Reach for six before you widen into a second category.

Staying narrow is not modesty, it is how sourcing works. Of all the category families our taxonomy applies, only 15 currently hold enough qualified creators to make filtering on them worthwhile, so brands shop from a short list of well populated lanes. Being the person who clearly belongs in one of them beats being adjacent to five.

While you are building the three, put them somewhere with a plain address: 67.9% of qualified profiles give a buyer nowhere at all to go and watch the work, so a bare page with three embedded clips and an email on it is already ahead of the field.

The send log, and the five columns that make it worth keeping

Outreach without a record is just mood. After twenty sends you want to know which of your three pieces travels, which product categories answer, and whether you changed anything between attempt six and attempt seventeen. A spreadsheet with the columns below turns a month of silence into information instead of a verdict on you.

One discipline makes it work: never send the same message twice without altering exactly one variable. Change the piece you lead with, or the category, or the opening line, but not two at once, or you learn nothing from the reply you eventually get.

Pricing the first job so it does not cap the next twenty

Quote per deliverable, never per hour, and name the permission you are selling in the same breath. Our published 2026 planning estimates put a fifteen to thirty second video from a nano creator at about $150, with a sensible band of roughly $100 to $250, and those are planning figures for a conversation rather than a going rate.

On a first job, sit at the bottom of that band if you must, but hold the add-ons at full price: raw footage around $50, an extra hook variation around $30, three months of paid usage around $40 for a nano creator. Discount the number of revision rounds instead, because revisions cost you time while usage rights cost you future income. A brand that gets twelve months of paid usage bundled free into a $120 job has bought your whole year at that product.

Accept gifting only when the retail value of the product clearly exceeds what the shoot costs you in time, and only for a single short piece with no paid usage attached. Put the scope in writing before you film: deliverables, delivery date, revision rounds, where the footage may run and for how long. Ask for half the fee up front on a first job with a brand you do not know, and give a payment deadline on the invoice rather than leaving it open.