Do not price Q4 on Sora API
Any rate card that assumes Sora API after 24 September 2026 is fiction. Rebuild quotes now.
Dead endpoints make dead margins. Any rate card that assumes Sora API availability after 24 September 2026 is fiction. If your Q4 packages still bake in that assumption, rebuild the quotes now while you can renegotiate like an adult instead of apologizing in October.
This is not panic. It is hygiene. Tools change. Contracts that pretend otherwise transfer risk to your sleep schedule.
Find the fiction in your current quotes
Open every active proposal that mentions generative video for Q4. Highlight lines that name Sora as a dependency, even soft ones like "or equivalent." Soft fiction still fails when a client heard Sora on a sales call and priced their expectations accordingly.
Then separate deliverables from engines. Deliverables are shippable assets with acceptance tests. Engines are how you make them. Clients buy deliverables. You buy engines with credits, labor, and time. When an engine dies, deliverables can still ship if you remapped jobs. If you sold the engine by name, you sold a liability.
Use the Sora API sunset migration plan as the operational backbone, then translate it into commercial language: what changes, what stays, what needs a change order. Silence is not neutrality. Silence becomes a surprise invoice argument.
List packages by exposure. High exposure: weekly volumes that assumed one-call storyboards. Medium: hero spots with flexible tools. Low: edit-heavy work that only used Sora for occasional B-roll. Attack high exposure first.
Rebuild cost from jobs and keep rates
Price the job sheet, not the brand name. For each deliverable, estimate:
- Credits per attempt on the mapped live row.
- Expected keep rate (keeps divided by attempts).
- Editor minutes to stitch, caption, and mix.
- Revision rounds included.
If you need a refresher on unit economics, read AI video cost per second with audio 2026 and keep your own measured keep rates beside any blog benchmarks. Your keep rate is the truth. Blog averages are a starting rumor.
Agencies that want package architecture language can cross-check AI video agency pricing models explained and then still do the unromantic math. Retainers without credit assumptions are how surprise overages become relationship damage. Project fees without revision caps are how "one more storyboard panel" eats December.
Build a buffer for migration learning. The first two weeks on a new row are educational. Education is billable if you planned it, charitable if you did not. Charitable work in Q4 is how studios start January tired and underpaid.
Talk to clients before they feel it
Send a short, calm note. State the sunset fact. State what deliverables remain unchanged. State what process changes. Offer a revised rate card or an addendum. Invite questions. Do not bury the note in a newsletter they will skip.
If a client demands the old named engine, offer a museum option only if you still have archived finals. New generation needs live rows. Be kind and clear. Clarity beats fake reassurance.
Sales teams should update one-pagers the same week. Old PDFs circulating in inboxes are how fiction repopulates. Version your decks. Kill the old link.
Protect Q4 cash without sounding cold
You can be warm and still firm. "We rebuilt this so your campaigns keep shipping" is a human sentence. "Not our problem, tool died" is accurate and unnecessary. Take responsibility for the remap. Do not take responsibility for keeping a dead API alive.
Offer phased migrations for large retainers: critical ads first, evergreen second, experimental third. Phase plans reduce panic and spread credit spend. Panic buying of attempts destroys keep rates.
When your commercial sheet and production sheet finally match, create a Versely account and run proofs on the rows you plan to sell so quotes reference reality. Dead endpoints make dead margins. Live jobs make boring, healthy Q4s.
Internal kickoff meetings should include finance, not only creatives. Finance needs to see credit burn scenarios at low, medium, and high keep rates. Creatives need to see which packages are no longer profitable under honest math. Shared screens prevent heroic guessing.
If you sell through partners or freelancers, update their briefing docs too. A partner quoting old assumptions can undercut your repaired rate card and then escalate the mess back to you when generation fails. One commercial source of truth is part of migration.
Track margin weekly in Q4 on migrated packages. If keep rates beat your buffer, you can share efficiency as bonus rounds or healthier profit. If keep rates miss, you adjust process before December. Waiting for a quarterly autopsy is how surprises compound.
Treat this as a weekly habit, not a one-time document. Reopen the checklist every Monday during busy seasons and ask what changed in tools, keep rates, and client expectations. Small updates prevent large apologies.
Write acceptance tests in language a new freelancer can run without a meeting. If a test needs tribal knowledge, it will fail under deadline pressure. Deadline pressure is when quality systems earn their keep.
Keep a short decision log: date, choice, reason, owner. Future arguments about taste become shorter when the log shows why a path was chosen. Logs are kindness to future you.
Separate exploration credits from delivery credits in your budget. Exploration is how you learn. Delivery is how you ship. Mixing them hides the true cost of both and makes every experiment feel like a production failure.
Phone-first review remains non-negotiable. If it only works on a calibrated desktop display, it does not work for your audience. Most of your viewers will never see your timeline.
When stakeholders disagree, return to the promise in the brief. The brief is the contract with the viewer. Tools are temporary. Promises are what retention measures.
Template what you can. Templating is not creative death. It is how calm teams ship on Thursday without heroic weekends. Save heroism for the rare idea that truly needs it.
Measure one primary metric per change. If you change the open, watch early retention. If you change the mix, watch completion and support tickets about clarity. Multi-metric celebrations hide what actually moved.
Teach the why in onboarding docs, not only the clicks. People abandon processes they think are bureaucracy. People keep processes that protect them from rework.
If a vendor sunset is involved, keep the primary source notice beside your internal plan. Secondhand summaries drift. Primary sources plus your job sheet beat rumor threads.
FAQ
What if a signed SOW names Sora explicitly?
Talk to your lawyer and your client early. Many SOWs allow equivalent tools for the same deliverable. Some do not. Early conversation beats silent substitution.
Should I raise prices or only change engine language?
Often both. New stitch labor and learning buffers are real costs. Pretending migration is free trains clients to expect miracles.
Can I wait until September 23 to revise quotes?
You can. You should not. Calendar proximity reduces leverage and increases apology volume. Rebuild now.