Back Office · office.temerarii.xyz
One asset, all the way in — composition, the wireframe + storyboard, the output format stack, and the template, all read from the SAME content-index record. The expected output matches what /media surfaces for this post.
post longform-W51-Wedkind longformweek W51date 2026-12-23campaign longform-youtubepillar emerging_techbeat asset videoduration 160.7sground blackscenes 8

Checklist the per-video bar — engine/sim

98.4/100
plain languagevo coverageno dead airuniquenesscaption fitcompletenesscleanliness
quantitative quality · weights learn from your reviews (engine.sim.memory review longform-W51-Wed good|bad)
⚠ 1 flag(s) — not yet ship-ready: copy_generic · see docs/strategy/VIDEO-CHECKLIST.md

Composition comp · template family · expected output

composition LongFormChaptersfamily / template LongFormChapters
9:16 Reelpending1:1 Squarepending16:9 Widepending9:16 4Kpending1:1 4Kpending16:9 4KpendingGIF (SMS)pending
render pending — silent master not yet on disk
expected output: 0/7 rendered — same matrix the /media preview surfaces for this asset.

Composition layer × scene 8 scenes · 160.7s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d
open0–23.4sspatial-parallaxicon·liquid-chrome♪ bed_in
The week is about AI and data fluency for going to market, and today is growth and performance marketing: the paid ads and funnels that bring leads and sales. Most teams hand this to one specialist and pray. By the end of this you will know how to train your own people to run and read a paid funnel with a model beside them.
on-screen: Train the team to run growth
expected on screen: black ground · cone hero in the shared Signal Field · Augur leads · node-graph · spatial-parallax · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape coneground blacktreatment liquid-chromemotion spatial-parallaxpower summoninstrument summon→Train the team to run growth
2s2
matches intent
NumberedList
template
teach23.4–43.9skinetic-buildicon·wireframe♪ node_lock
First move: write the funnel down in plain language. Someone sees an ad, clicks, lands on a page, fills a form, books a call. Five steps, five sentences. The team that can say the funnel out loud can find the leak in it. This is the lesson before any tool: name every step a customer takes.
on-screen: Map the funnel in plain words
expected on screen: black ground · a NumberedList panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id NumberedListvisual node-graphshape coneground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Map the funnel in plain wordscurate items, nodes
3s3
matches intent
ChecklistCard
template
teach43.9–64.4skinetic-buildicon·wireframe♪ node_lock
Second move: connect the model to your ad account through an adapter and ask which ads bring clicks that turn into booked calls, not just clicks. It reads the real spend and answers. The team learns to judge ads by the money at the end, not the likes at the top. That is the whole game.
on-screen: Pull the real ad numbers
expected on screen: black ground · a ChecklistCard panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ChecklistCardvisual node-graphshape coneground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Pull the real ad numberscurate items, nodes
4s4
matches intent
FlowSchematic
template
teach64.4–84.60000000000001skinetic-buildicon·wireframe♪ node_lock
Third move: take the worst step in the funnel and only fix that one. If people click but never fill the form, you ask the model to rewrite the form page using the words from your winning ads. One step, one fix. The team learns to mend the leak instead of pouring in more spend.
on-screen: Fix the weakest step
expected on screen: black ground · a FlowSchematic panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id FlowSchematicvisual node-graphshape coneground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Fix the weakest stepcurate nodes, stages
5s5
matches intent
AnnotatedDiagram
template
teach84.6–104.8skinetic-buildicon·wireframe♪ node_lock
Fourth move: ask the model to write the next ad to test, based on what already worked, and write down what you expect to happen. Then run it small. You are teaching the team to test on purpose with a guess written down first, so they actually learn from the result instead of just reacting.
on-screen: Let it draft the next test
expected on screen: black ground · a AnnotatedDiagram panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id AnnotatedDiagramvisual node-graphshape coneground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Let it draft the next testcurate callouts, nodes
6s6
matches intent
WireframeMock
template
teach104.8–125.3skinetic-buildicon·wireframe♪ node_lock
Honest proof: we run our own funnels with this exact loop, and we teach it the way we use it. No invented win rates here. The point is the method: name the funnel, read the money, fix one step, test the next. A team that can do that does not need to pray over a specialist.
on-screen: We point this at ourselves
expected on screen: black ground · a WireframeMock panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id WireframeMockvisual node-graphshape coneground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→We point this at ourselvescurate nodes
7s7
matches intent
RankList
template
proof125.3–145.1sreceipts-counticon·wireframe♪ node_lock
Wait, one more thing on the spend. When the model finds a leak, the first instinct is to add budget. Train your team to do the opposite: fix the step, then spend. A clean funnel makes every dollar work harder. That restraint is a skill, and it is the one most teams never learn.
on-screen: Mend the leak, not the budget
expected on screen: black ground · a RankList panel over a dimmed Signal Field · Augur leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id RankListvisual receiptsshape coneground blacktreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Mend the leak, not the budgetcurate rows, statsLabels
8s8
matches intent
shared field
signature-3d
resolve145.1–160.7scoalescenceicon·liquid-chrome♪ bed_out
So the takeaway: map the funnel, read the real money, fix the weakest step, test the next one, and mend before you spend. That is a team that runs growth instead of guessing at it. Watch us run live funnels at office.temerarii.xyz.
on-screen: Run a real funnel with us
expected on screen: black ground · cone hero in the shared Signal Field · Augur leads · coalescence · coalescence · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape coneground blacktreatment liquid-chromemotion coalescencepower coalescenceinstrument coalescence→Run a real funnel with us

Format stack 1 aspects · same scenes[], re-cropped

16:9
1920×1080
X/Twitter · YouTube · LinkedIn video

Channels 2 destinations

YouTubeBlog

Social captions supplemental published copy · per channel (comp_id level)

youtubeTrain your own team to run a paid funnel with AI beside them Most teams hand growth to one specialist and pray. By the end of this video you will know how to train your own people to run and read a paid funnel with a model beside them. The moves: - Map the funnel in plain words. Write the funnel down: someone sees an ad, clicks, lands on a page, fills a form, books a call. Five steps, five sentences. A team that can say the funnel out loud can find the leak in it. - Pull the real ad numbers. Connect the model to your ad account through an adapter and ask which ads bring clicks that turn into booked calls, not just clicks. It reads the real spend and answers. Judge ads by the money at the end, not the likes at the top. - Fix the weakest step. Take the worst step in the funnel and only fix that one. If people click but never fill the form, ask the model to rewrite the form page using the words from your winning ads. Mend the leak instead of pouring in more spend. - Let it draft the next test. Ask the model to write the next ad to test, based on what already worked, and write down what you expect to happen. Then run it small. Test on purpose with a guess written down first, so you learn from the result instead of just reacting. One caution on spend: when the model finds a leak, the first instinct is to add budget. Do the opposite, fix the step, then spend. A clean funnel makes every dollar work harder. That restraint is the skill most teams never learn. The honest proof: we run our own funnels with this exact loop, and we teach it the way we use it. No invented win rates. The point is the method: name the funnel, read the money, fix one step, test the next. Watch us run live funnels at office.temerarii.xyz Keywords: paid funnel, growth marketing training, ad account analytics, fix the weakest step, A/B test with AI, mend before you spend

Cross-links every lens is a view on this one record