Optimize-or-Originate Creative Portfolio
Separate scalable improvement from rare timeline-changing originality
- Difficulty
- Advanced
- Time to result
- ~ongoing to results
- Steps
- 6
- Confidence
- 91%
The Optimize-or-Originate Creative Portfolio divides content into two fundamentally different lanes. The optimization lane produces better versions and variations of known patterns, where AI, historical data, and performance metrics can create significant value at scale. The origination lane seeks the rare idea that is unpredictable from existing conventions and establishes a new creative timeline that others subsequently imitate. Because such ideas are uncommon, the framework does not demand originality from every asset. Instead, it prevents incremental metrics from consuming all creative capacity. Teams assign work to a lane before production, use efficiency and response data to judge optimization, and apply broader human judgments about culture, surprise, memorability, and strategic fit to origination. The portfolio recognizes that ordinary improvement drives results while a small protected frontier preserves the possibility of a breakthrough.
Origin
Extracted from the closing Marketing Against The Grain discussion about AI's strength at producing variations and humanity's continuing advantage in creating unpredictable new creative timelines.
Core principles
- 01Most useful content improves existing patterns rather than inventing new ones
- 02Rare original ideas create new timelines for others to follow
- 03Optimization and origination require different evaluation methods
- 04Use AI heavily for scalable variation while protecting human-led exploration
How to run it
- 1
Declare the Lane
Before production, classify the brief as optimization of an established pattern or an attempt to originate a new one.
Pro tip Use different briefs and review meetings for the two lanes.
Watch out Calling every campaign innovative makes the distinction meaningless.
- 2
Scale Optimization
For established formats, use AI, historical evidence, rapid variations, and campaign metrics to improve relevance, quality, and efficiency.
Pro tip Automate repeatable production while retaining brand-quality checks.
Watch out Incremental success can create fatigue if the same pattern is exploited indefinitely.
- 3
Protect Origination
Reserve time and resources for human-led exploration that is not required to resemble historical winners.
Pro tip Use cultural observation, contradiction, nostalgia, and changing human behavior as exploratory inputs.
Watch out Immediate efficiency thresholds can kill original concepts before they have a chance to work.
- 4
Use Lane-Specific Evaluation
Judge optimization through repeatable performance evidence; judge origination through strategic fit, surprise, cultural relevance, and informed human conviction.
Pro tip Define the acceptable downside of experiments in advance.
Watch out Do not use a single scorecard for both predictable improvements and genuine creative bets.
- 5
Recognize a New Timeline
When an original concept succeeds and begins inspiring variations, convert its repeatable elements into the optimization lane.
Pro tip Document why the concept was different before imitation obscures its original mechanism.
Watch out Copying the surface form without understanding the idea quickly degrades its value.
- 6
Rebalance Continuously
Adjust the allocation between reliable output and exploratory work according to brand maturity, campaign needs, and available creative talent.
Pro tip Protect at least some exploratory capacity during periods of aggressive performance optimization.
Watch out An all-origination portfolio is unstable, while an all-optimization portfolio eventually becomes indistinguishable.
In the wild
A successful original advertisement is followed by roughly two years of variations as other marketers adapt its pattern. Eventually, another unpredictable advertisement establishes a different timeline and becomes the new source of imitation.
→ The rare original asset creates a category pattern, while the larger optimization market scales variations of it.
A CPG brand uses AI to produce many higher-quality versions of proven product-ad formats while reserving a smaller human-led program for culturally surprising concepts that may not resemble past winners.
→ The brand gains reliable output without abandoning the possibility of category-defining creativity.
Common mistakes
Demanding Originality Everywhere
Most useful content is not category-defining, and requiring every asset to be unprecedented wastes resources and destabilizes production.
Letting Metrics Own Both Lanes
Tightly clustered short-term results can identify incremental winners while suppressing culturally significant but less predictable ideas.
Eliminating the Frontier
A portfolio composed entirely of optimized variations eventually becomes part of the sameness audiences learn to ignore.
Is it for you?
Best for
It is best for brands that need dependable content volume while still seeking occasional culturally distinctive campaigns.
Not ideal for
It is not ideal for teams without enough capacity to maintain both an optimization engine and a protected experimental lane.
From the transcript
“you have a successful ad and then you have maybe two years of variations of that ad.”
“And then someone comes out with something that creates a whole new timeline of ads because that ad is so original and unpredictable from what…”
“I think um originality is a very small part of what exists in the content space for a reason because it's really hard to be…”
From the episode
This AI Startup Is Beating Apple and NVIDIA at Their Own Game