Hi everyone. I've been looking into Sora for generating video assets for some internal reports and social media clips at my company. My boss is asking for a budget proposal, and the big question is whether switching to generated video from traditional stock footage subscriptions would actually save us money.
I'm trying to build a realistic annual cost breakdown, but I'm hitting some unknowns. For a traditional stock site, let's say we use 2-3 clips per month. A standard subscription for HD footage might be roughly $800/year. That gives us a known library, predictable quality, and specific legal coverage.
With Sora (assuming the rumored pricing models), the cost seems tied purely to usage. If we generate 25 clips a month, and the cost is in the ballpark of per-second or per-clip... the math gets fuzzy fast. My biggest unknowns are:
1. **Iteration cost:** How many generations does it take to get a usable clip? Each one adds up.
2. **Editing needs:** Will Sora clips require more work in a video editor to fit perfectly, adding labor time?
Here's a super simplistic comparison I sketched out in SQL to model different scenarios, but I need better inputs:
```sql
WITH traditional AS (
SELECT 800 as annual_subscription_cost,
30 as estimated_clips_per_year
),
sora AS (
SELECT 0.05 as assumed_cost_per_clip, -- This is a complete guess!
25 as clips_per_month,
12 as months,
3 as average_generations_per_final_clip
)
SELECT
t.annual_subscription_cost as traditional_cost,
(s.assumed_cost_per_clip * s.clips_per_month * s.months * s.average_generations_per_final_clip) as estimated_sora_cost
FROM traditional t, sora s;
```
Has anyone done a practical, real-world comparison for a similar use case? I'm especially curious about the hidden costs—like the time spent crafting the perfect prompt versus searching a stock site. For a team that uses about 30 short clips a year, is Sora already a viable alternative, or is the cost and consistency still in favor of traditional footage?