Jerry Novick

The point is diagnostic: if the hook fails, the concept or opening frame needs reworking, not the offer, because nothing downstream matters if nobody stops. They put a working number on it, calling a thumb-stop or hook rate of 30% or above good, and treat the elements of a hook as separate levers a brand can test rather than one impression of overall polish.

A Scribe ad made by the creator group Savvy Finds stopped them before the video even played for two reasons: the person on screen could plausibly have been the viewer, and the overlay text "What my boss sees" opened a question that had to be resolved by watching.

• Both elements had to work together.
• By the time viewers registered that the ad was about documentation software they were not shopping for, they were already through the hook.

Read original · Original 2

A hook rate of 30% or above is good, while a mid-to-high 20% result is worth examining if other metrics look promising.

• Hold rate should be above 20%; a strong hook and weak hold point to a disconnect where the body does not pay off the promise.
• Click-through rate should be 2% or above, read alongside hook and hold to separate a creative problem from a targeting problem.
• Conversion rate should be 4% or higher, though 2–3% can still be promising on high-ticket items.
• Return on ad spend should be above 2.5, ideally above 3.0 for serious scaling, though high-lifetime-value subscriptions and high-margin products can tolerate less.
• Frequency combined with thumb-stop ratio over time flags creative fatigue; a reading above 3 means it is time to refresh or rotate.
• Their distinctive decision rule: iterate on concepts where hook and hold rates are strong but conversion or return underperform, because those are high-signal assets with fixable gaps, and cut concepts where the hook fails, whatever happens downstream.

For a subscription product with high lifetime value, they say a return on ad spend below 1.0 can still be acceptable, and a conversion rate between 2% and 3% can show promise on expensive items. The exceptions are built into the thresholds rather than left to judgement.

Read original

Early Meta copywriting for a supplements brand produced strong click-through, low acquisition cost and strong return, then fell off sharply.

• The cause was found after the click, in the landing experience, not in the ad.
• Their conclusion: when performance tanks, it is rarely the ad; it is the system.
• They connect that lesson to Meta's platform changes: algorithms evolve, but systems endure.
• The practical consequence is that a creative brief should cover the post-click path — bridge pages, education steps and message match between ad and landing page — not just ad units.

Ads were sending cold traffic to a product page on a crowded site with a prominent buy box at the top.

• Visitors hit sticker shock, browsed other products and left.
• They built a multi-step, education-led funnel with a bridge page written in the voice of the brand's doctor spokesperson, so people arrived at the product page already informed.
• Performance recovered within a week and exceeded earlier levels.

Read original

Their added argument is about where the method strains.

• Moving from a product ad to a service business required reworking a prompt built for ecommerce, and compliance constraints on claims meant rewriting every word the AI produced.
• Cloning starts a testable variant; it does not ship a finished one.
• Categories with regulated copy should budget for a full human rewrite.

They deliberately took the exercise off its intended track, testing whether a direct-to-consumer product ad could translate into a service-provider ad.

• Finding an ad that would survive the translation took real digging.
• The final asset is described as proprietary and unlaunched, so it cannot be inspected.
• The guest instructor's method is Will Sartorius's, taught in a paid bootcamp, and the prompt itself was withheld as course material.

They deliberately departed from the instructor's direct-to-consumer framing, naming where the taught method needed modifying rather than presenting the result as a straight application.

Read original

The argument is less a product pitch than a claim about what the role now rewards: the analyst, creator and organiser parts of the work are increasingly tooled, while judgment lies in what you do with the output.

They name Motion (Creative Analytics) as the tool, describe joining its eight-week bootcamp and posting the weekly homework publicly.

• Week one covered hooks with instructor Evan Lee, plus a group coaching session.
• They credit paid media managers Eric Santiago and Madison West for the account-side work.

Read original · Original 2 · Original 3

Start here for the basic vocabulary and a worked example before reading about campaigns.

Novick defines a hook as whatever stops a scroll in an ad: the image, a text overlay or the first spoken words.

• Video hooks are generally judged on the first three seconds.
• They put a rough number on it, saying a good hook or thumb-stop rate is 30 percent or above.
• They then take apart a specific ad by the creator group Savvy Finds for the software Scribe, naming two things that stopped them before the video even played: a person on screen who could plausibly be the viewer, and the text overlay "What my boss sees," which opens a question the viewer has to resolve by watching.
• The two hooks work together, carrying the viewer past the moment they realise this is an ad for something they were not looking for.
• The post says the full ad image is in the first comment.

For brands planning a campaign

Useful for triaging DTC creative: evaluate the first three seconds as their own problem, separate from the offer and product, and consider pairing a recognisable on-screen person with an open-loop text overlay.

This is the most directly reusable piece in the set for anyone running paid social.

Novick lists the metrics they prioritise for direct-to-consumer advertising and gives working thresholds: a hook rate of 30 percent or above, hold rate above 20 percent, CTR of 2 percent or above (3 percent and up gets special attention), CVR of 4 percent or higher, and ROAS above 2.5, ideally above 3 for serious scaling.

• They also flag frequency and thumb-stop ratio over time as signals that creative is fatiguing; a reading above 3 means it is time to refresh or rotate.
• The thresholds come with stated exceptions: subscription products with high lifetime value can tolerate ROAS below 1.0, and a CVR between 2 and 3 percent can still show promise on high-ticket items.
• The decision rule is explicit: iterate on concepts where hook rate and hold rate are strong but CVR or ROAS underperform, and cut or deprioritise concepts where the hook fails regardless of what happens downstream.
• A graphic is credited to the Motion (Creative Analytics) website.

For brands planning a campaign

Gives a DTC team a diagnostic order: check hook and hold first to see whether the problem is creative or targeting, then decide whether to refresh, iterate or cut.

This is the clearest account of Novick's full-funnel thinking and worth reading before judging them on creative alone.

Novick describes early Meta ad work for the supplements brand Purity Products, where the copy delivered strong click-through and return on ad spend before performance fell off sharply.

• A paid media manager identified the cause: the ads sent cold traffic to a product page with a prominent buy box near the top, where visitors hit sticker shock, wandered into other products and left.
• The fix was a multi-step, education-led funnel with a bridge page written in the voice of the brand's doctor spokesperson, so people arrived at the product page already informed.
• Novick says performance recovered within a week and frames the lesson as "clever ads don't fix a broken system," connecting it to Meta's algorithm changes.
• The post teases a follow-up series on funnels, creative systems and message match.

For brands planning a campaign

Before rewriting ad copy, check the landing experience and message match. For a DTC brand, a cheaper or higher-priced entry bridge page with a credible spokesperson voice may recover performance that creative changes alone cannot.

This is the most honest account here of the rework involved.

Novick describes a session with Will Sartorius on adapting a competitor's high-performing ad for your own brand's voice, colours, fonts and benefits, using Claude to build a prompt and then an image tool to produce the asset.

• Rather than repeat the exercise as taught, Novick tested whether a direct-to-consumer product ad could translate into a service business.
• They say finding a suitable ad took real digging, the original prompt structure was too product-oriented and needed reworking, and compliance concerns led them to rewrite every word the AI generated.
• The resulting asset is proprietary to a specific service provider and unlaunched, while the prompt is withheld as bootcamp material.
• Ask to see a launched example and the prompt.

For brands planning a campaign

Treat competitor-ad cloning as a starting point rather than a finished asset. Budget for rewriting AI-generated copy, especially in regulated categories, and expect to rework the prompt when the offer is a service rather than a product.

Useful background if you are deciding whether to work with them.

Novick contrasts the older shape of the creative strategist job—working inside Meta accounts alongside paid media managers and managing charts, spreadsheets, file naming conventions and sprawling folders—with an approach built around software for creative analysis, performance tracking, research, AI-assisted copy and ad account organisation.

• They name Motion (Creative Analytics) as the tool and say they joined its eight-week bootcamp, with week one covering hooks with Evan Lee and a group coaching session, before posting their homework.
• They credit two paid media managers, Eric Santiago and Madison West, for the account-side work.

For brands planning a campaign

For a role involving creative analytics tooling, expect someone who leans on those tools and partners with paid media managers. Ask what data they track and how they make creative decisions.

A thumb-stop rate of 30% or above is cited as a good benchmark.

• In a worked example, a Scribe ad from the content collective Savvy Finds stopped the thumb before playback for two reasons: the person on screen could have been the viewer, and the overlay text "What my boss sees" opened a mental loop that had to be closed.
• Novick argues that both elements had to work together, and that viewers processed the hook before registering that the ad was about SOP software.

Breaks an ad into its hook elements — casting or image, overlay text, first spoken line — and explains the psychology each triggers instead of judging the ad as a whole.

For brands planning a campaign

For a DTC brief, ask this creator to bring several candidate hooks and explain what makes each one stop the scroll. Test them against a 30% thumb-stop target rather than judging creative on polish alone, and ask for separate recommendations on casting and overlay copy.

Examples: Source 1 · Source 2

Hook rate: target 30% and above, with a mid-to-high 20% result worth examining if other KPIs look promising; a low hook rate points to the concept or opening frame rather than the offer.

• Hold rate: above 20%; a strong hook with weak hold signals that the body does not deliver on the promise.
• CTR: 2% or above, with 3% and higher getting special attention; read it alongside hook and hold to distinguish creative problems from targeting problems.
• CVR: 4% or higher, with 2–3% potentially promising on high-ticket items; strong CTR with collapsing CVR points downstream, sometimes to the offer or a creative-to-landing-page mismatch.
• ROAS: above 2.5, ideally above 3.0 for serious scaling, though high-LTV subscriptions and high-margin products can tolerate lower, even sub-1.0.
• Frequency combined with thumb-stop ratio over time flags creative fatigue; crossing above 3 means it is time to refresh or rotate.
• The decision rule: iterate on concepts with strong hook and hold rates but weak CVR or ROAS, because these are high-signal assets with fixable gaps.
• Cut or deprioritize concepts where the hook fails, regardless of downstream results.

Uses paired metrics to diagnose performance: hook and hold isolate the creative, CTR and CVR show whether the break is in the ad or downstream, and frequency helps determine when to act.

For brands planning a campaign

This is close to a ready-made scorecard for a DTC engagement. Set target thresholds and the refresh trigger in the brief, and ask for iteration plans that specify which metric justifies keeping, fixing, or killing a concept.

Examples: Source 1

The copy produced strong click-through rates, low cost per acquisition, and enviable ROAS, then performance dropped sharply.

• Working with a paid media manager, they located the cause after the click: traffic was being sent to a product page on a crowded site with a prominent buy box.
• Customers hit sticker shock, browsed other products, and left.
• Novick's first multi-step, education-driven funnel was the fix.
• It used a bridge page written in the voice of a doctor who served as the brand's spokesperson, so visitors arrived at the product page already warmed.
• Performance recovered and exceeded prior levels within a week.
• Novick's conclusion: when performance tanks, it is rarely the ad; it is the system.
• They apply that lesson to Meta's Andromeda update: algorithms evolve, but systems endure.
• A follow-up series is promised on funnel architecture, creative systems that teach the machine who to find, the art and math of message match, and case studies.

For brands planning a campaign

Scope a DTC engagement to include the post-click path, not just ad units. Ask this creator to diagnose the funnel and message match before producing new creative, and show how a bridge page or education step would carry the ad's promise into the product page.

Examples: Source 1

The process uses Claude to build a prompt, which is then run through NanoBanana.

• Novick deliberately took the method off its intended track, applying a product-ad clone to a service provider instead of a DTC product.
• That required modifying the DTC-oriented prompt structure and rewriting every word of the AI-generated copy because service categories can carry compliance constraints on claims.
• The output is proprietary to a specific service provider and was not published.
• Ask to see a sample prompt and final creative before commissioning the method.

Competitor ad as a structural template, prompt generation via Claude, image or video generation via NanoBanana, followed by a full human rewrite of copy for regulated or service contexts.

The prompt is taught inside a paid bootcamp and was not shared publicly. Ask for a sample prompt or finished creative before commissioning this workflow.

For brands planning a campaign

Competitor-inspired creative is presented as a repeatable pipeline, not one-off inspiration, which suits brands seeking a volume of testable variants. For services or regulated categories, budget for a human rewrite and compliance pass over generated copy; do not expect a direct clone to ship.

Examples: Source 1

Previously, creative strategy meant working inside a Meta account, often alongside paid media managers, switching between charts, numbers, and multiple screens, and combining the work of analyst, creator, and — in their phrasing — witch doctor.

• The job also involved spreadsheets, project management software, file naming conventions, and many folders, with the recurring risk of things falling through the cracks.
• Motion (Creative Analytics) is credited with changing that through creative analysis, performance tracking, research, AI-driven copy creation assistance, and ad account organization.
• Novick joined Motion's eight-week bootcamp, completed week 1 — a hooks workshop with Evan Lee and a group coaching session — and has posted the weekly homework assignments publicly, including the Scribe hook breakdown and the week 4 cloning exercise.

For brands planning a campaign

Signals a research- and tooling-led process: a brand can expect creative proposals informed by ad-library research, performance analytics, and structured analysis rather than intuition alone. Part of this creator's process is already public, giving brands a way to assess their working style.

Examples: Source 1 · Source 2 · Source 3

Each takes a single assignment or question — what a hook is, which metrics matter, how to clone a competitor ad — and answers it with a named benchmark, a concrete example, and a brief explanation of why the example works.

• The Scribe ad is analyzed frame by frame; the metric post is a bulleted threshold list; the funnel story uses a before-and-after narrative.
• Novick also names collaborators, instructors, and tools, and previews what is coming next.

For brands planning a campaign

This creator documents reasoning publicly in small units, giving a brand a low-cost way to sample the working method before commissioning anything: read a teardown, then ask for the same treatment of the brand's own creative.

Examples: Source 1 · Source 2 · Source 3 · Source 4

The post does not give a publication date.

View example

They praise Motion’s tools and training; ask whether the arrangement was paid or otherwise commercial.

View example

They do not describe a business relationship with Scribe.

View example

They do not describe a business relationship with the group.

View example

They ask why an ad worked, what the account data shows and what to test next. Both write for DTC and performance-marketing readers, treating creative strategy as a discipline with its own practice and vocabulary rather than a subset of media buying.

Papon's posts make the case for the role and reflect on Cannes Lions, highlighting the rooms and people worth being around.

• Novick publishes step-by-step lessons with named benchmarks, worked ad breakdowns and explicit decision rules.
• Readers looking for the argument for the role should start with Papon; those looking for the method should start with Novick.

Read original · Original 2

Their post on editing lays out a direct-response structure — stop the scroll with a hook, introduce the problem, make the value obvious, build trust, use social proof to cut scepticism, handle objections, close — that maps closely onto Novick's hook-then-body-then-conversion framing. Both warn against volume as a substitute for quality: cheaper production and easier variation do not make people care.

Ganzina focuses on the craft of video and the editing decisions that support each stage of an ad, including cutting a beautiful opening shot that takes four seconds to make its point. Novick focuses on diagnosis and strategy: which metric tells you the hook failed, when to refresh creative, and why the landing experience, not the ad, may be the problem.

Read original · Original 2

Both start with hook and hold performance, and argue that specific, research-led copy beats copy aimed at everyone.

Malik is a conversion copywriter whose case studies centre on landing pages, email sequences and client ad copy, with results presented as before-and-after outcomes.

• Novick is a creative strategist who works on the whole creative-plus-funnel system.
• They publish a single metric hierarchy with thresholds and an iteration rule rather than client results, and teach mostly by taking apart other people's ads.

Read original · Original 2

Their examples describe editing around a specific job: get attention, make the product’s value clear, build trust, answer objections and give viewers a reason to act.

• One example is cutting a polished opening that takes four seconds to get to the point; another is keeping an unpolished customer quote because proof matters more than aesthetics.
• Nikita is the more focused choice when the brand needs an editor who will prioritize conversion over polish.

The posts focus on ad structure and editing, not diagnosing hook and funnel metrics or rebuilding the post-click journey. Jerry Novick’s posts are a closer match if the problem may extend from the ad into the landing experience.

Examples: Source 1 · Source 2

Their examples argue for writing to one specific buyer and show how audience research can shape the message: one case describes using audience language from Reddit in a page rewrite, alongside a reported change in conversion rate from 2.1% to 9.7%. The examples also cover email and ad work, making Amir a useful alternative when a campaign needs broader conversion copy rather than a creative-and-funnel diagnosis centered on Meta ads.

The posts do not explain a hook-and-hold testing system or an ad-to-product-page funnel comparable to Novick’s Purity Products account. Amir presents the reported results, which the posts do not independently verify.

Examples: Source 1 · Source 2

Their post frames each ad as a chance to learn about performance, buyer behavior and creative patterns, which may suit a brand seeking an ongoing testing mindset rather than a one-off copy or editing assignment.

The posts make a broad case for creative strategy but provide fewer concrete examples of how Lauren diagnoses a specific ad or repairs a post-click funnel. For a proposed Meta ads test, Novick’s hook breakdown, metric thresholds and funnel account offer more detailed starting points.

Examples: Source 1