@JadeJoker solid point - cadence matters more than volume. What moved the needle for me was running micro tests weekly on one variable only, hook vs first frame vs caption CTA, same posting window so signal isnt muddy - track saves + 2s retention not just likes. Once a format hits baseline I lock it in + scale reps, everything else gets nuked fast so the feed stays clean.
From my experience, regular testing is really helpful. Even small experiments (new hooks, video length, captions, posting times, or formats) every 1–2 weeks can show what your audience actually responds to. Platforms change often, so what worked before may not work now. Consistent testing makes it easier to spot patterns, double down on what performs well, and drop what doesn’t.
When scaling companies, by what percentage do you increase the budget?
Most people I know test small variations every 1–2 weeks (hooks, length, format, captions, posting time). Regular testing makes patterns clear over time – what gets better retention, saves, or comments. Instead of guessing, you slowly double down on the formats that consistently perform and drop the ones that don’t.
I try to test new content styles regularly, but in a controlled way. Usually I keep a core format that performs well and then experiment with small variations alongside it. That way I’m not constantly resetting performance, but I can still spot new opportunities. Over time, consistent testing has definitely helped me understand what resonates organically instead of relying on assumptions.
Normally as a Meta suggestion. But if it's too high, we just scale 10-20%
Thanks for the answer, I usually increase by 20-30% and observe
I try to test something new every few weeks, whether it’s a different video style, hook, posting time, or content angle. Regular testing definitely helps, because platforms and audiences change fast. Keeping track of what gets higher watch time, saves, and comments makes it easier to double down on formats that actually resonate and drop the ones that don’t.
