Blog Playbooks · The Benchmark Playbooks · Part 9
The phone-shot UGC playbook: why the cheap-looking ad keeps winning
Phone-shot ads became winners at more than twice the rate of high-production ads in our data, and at a lower cost per lead. The group is small. Here is how to test it without a crew or a budget.
We have run over 18,920 Meta ads, and one of the clearest splits in the data is about how an ad was made. When we labeled 2,126 ads by production style, phone-shot UGC (user-generated content, or anything that looks like it) became a winner 15.5% of the time. High-production ads, the polished kind with lighting, edits and a crew, became winners 7.0% of the time. Phone-shot ads also brought leads cheaper: $22.21 each, against $25.50 for high production.
The catch is size. Only 64 ads in our labeled set were phone-shot UGC. High production had 784, the biggest production group by far. So the polished ad is what most accounts make, and the phone-shot ad is what most accounts should be testing more of.
The numbers, both ways
A winner, in our data, is an ad that spent at least 10 times its own account's median ad, and at least $500. In plain words, it is an ad Meta's system chose to put real money behind.
Source: Creative Benchmarks, benchmark.sneeze.it (2,126 labeled ads)
On cost per lead, the order is close but not identical: phone-shot UGC $22.21, high production $25.50, text only $26.00, lifestyle photo with text $26.38, animation $31.63, mixed still and video $32.88. The labeled average across all 2,126 ads was $26.62.
Phone-shot UGC ranked first among production styles on both measures: the lowest cost per lead and the highest winner rate. That matters, because cost per lead and winner rate often disagree. An ad can be efficient with the little money it gets and still never get much. Phone-shot UGC did well on efficiency and on attracting spend. We explain why the two measures split in winner rate vs cost per lead.
Source: Creative Benchmarks
Why it might work, and what we cannot tell you
We can tell you what happened. We cannot prove why. Here is our best reading, offered as a reason to test, not as a finding.
A phone-shot video looks like something a friend posted. It sits naturally in a feed full of things friends posted. A polished ad looks like an ad, and people have learned to scroll past ads. A phone-shot video also tends to show a real person in a real place, which answers the question a local customer is actually asking: what is it like to walk in there?
What we cannot tell you is whether your phone-shot ads will match these numbers. Sixty-four ads is a small group. A few strong ads can lift a small group's rate a long way. We also cannot separate the style from the content: the people who shoot phone videos may also be saying different things in them. Treat 15.5% as a strong reason to run the test, and your own account's numbers as the answer.
The polished ad is what most accounts make. The phone-shot ad is what most accounts should be testing more of.
The playbook
- Pick who will be on camera
An owner, a staff member, or a customer who has given written permission. Someone who knows the place and can talk about it without a script. If a customer appears, get their written consent to use the video in ads before you film.
- Write three opening lines, not a script
Each one says what the viewer gets, in the first few seconds. Our playbook on how to open an ad covers which openings did best. Let the rest be natural.
- Shoot vertically on a phone, in the real place
Hold the phone upright, film where the customer will actually be, in normal light. Clean the lens. Make sure the voice is clear; sound matters more than picture.
- Keep the edit light
Trim the start and the end. Add captions, since many people watch with the sound off. Skip the stock music, the logo animation and the color grading.
- Launch three or four versions together
Different people, different openings, same offer. One video tells you about one video. Several tell you about the style.
- Run them beside your polished ads
Same campaign, same audience, same week. You want to see which ones Meta chooses to spend on when it has both to pick from.
- Read both numbers, then make more of what won
Look at where the spend went and at cost per lead. If phone-shot versions win on both, shoot more next month with new people and new openings. If they lose in your account, you have spent very little to learn it.
Make it a habit, not a one-off
The deeper lesson from the data is about volume. Across 97 accounts with 50 or more leads, the median account launching 4 or more new ads a week found 1.00 winner a month. Launching under 1 a week, the median was 0.37. The typical account launched 1.6 new ads a week. Phone-shot ads are the easiest way to raise that number, because they cost an afternoon instead of a production budget. A video you can make in ten minutes is a video you can make every week.
Here is the arithmetic. Say you spend $1,500 a month on lead ads. At the labeled average of $26.62, that buys roughly 56 leads. At the phone-shot UGC average of $22.21, it would buy roughly 67. That is not a forecast for your account. It shows that even a modest difference in cost per lead adds up over a month, before you count the higher chance of finding a winner. For more on launch rhythm, read how many new ads a week.
