Ad Performance Metrics: The DTC Creative Team's Guide

Ad performance is often judged by the numbers closest to the cash register. CTR looks healthy, ROAS clears the target, and the ad gets called a winner. That advice is incomplete. A high CTR can come from a curiosity hook that attracts people who never wanted the product, while strong ROAS can reflect a warm audience, favorable attribution, or cheap delivery rather than strong creative.

The better question is not, “Did this ad get clicks?” It's, “Which part of the creative caused the result?” Hook rate tells you whether the opening stops the scroll. Hold rate tells you whether the promise survives beyond the opening. Frequency tells you whether the same audience is seeing the concept too often. Conversion metrics then show whether the message and offer can turn attention into business.

That distinction matters for DTC teams running high-volume paid social. If you optimize media efficiency while ignoring creative signal, you can spend weeks changing audiences, bids, and budgets when the actual problem is a weak first frame or an unclear product benefit.

Why Most Teams Read the Wrong Metrics

CTR and ROAS are useful, but they're often treated as verdicts when they're really outcomes. CTR measures response to the entire ad and its audience. ROAS measures attributed revenue against spend. Neither one tells you whether the concept earned attention, communicated the benefit, or created demand that would have existed without the ad.

That's why a campaign can produce a respectable CTR while wasting production effort. A sharp headline might create enough curiosity to generate clicks, but the landing page then reveals that the product doesn't solve the problem implied by the hook. The ad wins the click and loses the customer. The metric that looked strongest was the one hiding the mismatch.

What is the difference between creative signal and media efficiency?

Creative signal answers questions about the asset:

• Does the first frame interrupt scrolling?
• Does the body deliver the opening promise?
• Does the value proposition make sense before the click?
• Does the call to action fit the level of awareness?

Media efficiency answers different questions:

• Did the auction find an affordable audience?
• Did the bid strategy deliver efficiently?
• Did placement and budget allocation support the result?
• Did the campaign reach people already close to purchase?

A strong ROAS result can therefore come from audience quality while the ad itself contributes very little incremental persuasion. The opposite can also happen. An ad may show strong hook and hold signals but weak purchase efficiency because the offer, product page, or checkout experience breaks the journey.

Practical rule: Read ad performance metrics as a chain. Diagnose attention first, sustained relevance second, action third, and revenue last.

A DTC creative review should therefore start at the asset level. Ask which opening, angle, proof point, and offer produced the signal. Only then should you decide whether to change the audience or scale the campaign.

A useful way to think about this distinction is to imagine two ads selling the same product to similar audiences. One ad gets cheap delivery because the platform can find likely buyers fast, but the creative itself says little beyond a generic claim. The other ad costs more to deliver, yet people stop, watch, and understand why the product matters. If you judge only by surface efficiency, you may back the first ad and underfund the second one, even though the second contains more reusable creative learning.

That mistake gets expensive over time. Teams often archive a concept because ROAS disappointed, when the real issue was a weak landing-page handoff. Or they scale a mediocre concept because it happened to run in a favorable pocket of demand. In both cases, the team learns the wrong lesson. The report says one thing, but the footage, copy, and audience behavior say another.

The goal is not to ignore business metrics. The goal is to put them in the right order. Attention metrics tell you whether the asset won the first battle. Click and conversion metrics tell you whether the message created intent that survived into action. Revenue metrics tell you whether that action translated into financial value. The closer you stay to that sequence, the easier it becomes to diagnose the actual cause.

The Core Ad Performance Metrics You Must Track

A useful measurement stack starts with basic delivery metrics, then follows the user through action and revenue. Each metric has a job. Treating every number as a creative KPI creates confusion.

Which delivery and cost metrics matter first?

Impressions count ad deliveries. They tell you how much exposure the platform generated, not whether people noticed the message. Reach counts distinct people, while impressions can include repeated exposure. Use impressions to understand delivery volume and use frequency to interpret how concentrated that delivery has become.

Historical measurement has moved in this direction before. Industry groups began formal work on viewability in 2007, created the Making Measurement Make Sense initiative in 2011, and reached a major milestone when the Media Rating Council lifted its viewable-impression advisory in the first half of 2014. The standard treated a display ad as viewable when at least 50% of its pixels were in view for at least 1 second, and a video ad as viewable when at least 50% was in view for at least 2 seconds, as documented in this history of viewability measurement. The lesson still applies: a served impression isn't the same as a real opportunity to see an ad. What this means for a DTC brand: treat your impression count as inventory the platform sold you, not as attention your creative earned, and let hook rate carry the attention question.

CPM is calculated as spend divided by impressions, multiplied by 1,000. It reflects the price of delivery in the auction. A rising CPM can point to competition, audience constraints, or fatigue, but it doesn't identify the creative problem by itself.

CPC is spend divided by clicks. Cheap clicks aren't automatically useful. Pair CPC with conversion rate, because low-cost traffic can still produce expensive acquisition when the post-click experience or audience intent is weak.

CPA is spend divided by conversions. It's closer to a business decision than CPC, but it still needs a conversion definition that remains consistent across campaigns.

These metrics sit near the top of most dashboards because they are easy to retrieve and easy to compare. That convenience is useful, but it can also create false confidence. Delivery and cost metrics describe what the platform accomplished in distribution. They do not automatically describe what the creative accomplished in persuasion.

For DTC teams, that distinction is important when reviewing early tests. In the first days of a launch, CPM may move for reasons that have nothing to do with the message. Audience size, bid approach, placement mix, and account history can all shape what it costs to buy attention. If the team treats those shifts as direct feedback on the concept, it may rewrite an ad that simply needs a fairer read.

A practical habit is to label each metric by job during reviews. Impressions show exposure. CPM shows the cost of access. CPC shows the cost of traffic. CPA shows the cost of a completed action. Once those roles are clear, the team stops asking one number to answer every question.

How should you read response and revenue metrics?

CTR is clicks divided by impressions. It indicates whether the ad motivates a click, but platform definitions vary. Some social reporting views include multiple types of clicks, while search reporting generally centers on clicks to the destination. Use one agreed definition in your scorecard before comparing variants.

CVR, or conversion rate, is conversions divided by clicks. It reveals whether the traffic generated by the ad can complete the intended action. A strong CTR with weak CVR usually deserves a message, offer, audience, or landing-page investigation, not immediate scaling.

ROAS is revenue divided by spend. It's a lagging efficiency measure, and it doesn't include every cost involved in producing or delivering the product. Use it for budget and business decisions, not as the only creative quality score.

For a deeper explanation of why cost metrics can point to different decisions, use this guide to CPM versus CPA. If you want the same stack wired into a working view rather than a spreadsheet, our page on ad performance analytics covers the tooling side. The practical standard is simple: never approve a creative from one metric alone.

Response and revenue metrics are where most debates become heated, because they sit close to business outcomes. That makes them essential, but it also makes them easy to misuse. When a team sees high CTR, people assume the ad must be strong. When they see healthy ROAS, people assume the concept deserves more investment. Both conclusions can be right, but neither conclusion is safe without context.

The main discipline is to ask what each metric can actually prove. CTR can prove that people were willing to take a next step. It cannot prove that the ad qualified the click well. CVR can prove that traffic turned into action at some rate. It cannot prove that the creative alone drove that action. ROAS can prove that attributed revenue exceeded spend by a certain ratio. It cannot prove that the ad was the true source of incremental demand.

That is why DTC teams benefit from separating performance review into two layers. The first layer asks whether the ad generated the intended action path. The second asks why that path happened. In practice, this means a strong CTR should trigger examination of the promise and click intent. A weak CVR should trigger examination of continuity between the ad and the destination. A strong ROAS should trigger examination of audience temperature before the team labels the creative a breakthrough.

Over time, the best teams become less impressed by isolated wins and more interested in repeatable patterns. They want to know whether a testimonial angle consistently lifts click quality, whether demonstrations consistently improve intent, or whether an offer-led approach creates conversion at the cost of long-term margin. That kind of learning comes from reading response and revenue metrics as evidence, not as final truth.

Creative-Level Diagnostics That Reveal Ad Quality

Cheap traffic can hide a weak creative. Read the ad in sequence, from the first frame to the post-click action, then add frequency to see whether the result holds as the audience sees it again.

Hook rate measures 3-second video views divided by impressions, multiplied by 100. It answers one focused question: did the opening earn enough attention to interrupt scrolling? Keep it apart from thumb-stop ratio, which is an internal diagnostic built on a shorter viewing threshold. A team that reports the two in one column compares numbers that were never measuring the same thing, which is why hook rate and hold rate get read against each other rather than merged. Motion puts a strong Meta hook rate at 30 to 40% (key metrics for creative performance). The kill bar we use at Selzee is tighter and carries a spend floor with it: retire an opening whose hook rate sits under 20% after roughly 500 impressions, because an opening that cannot earn the stop rarely recovers. What this means for a DTC brand: those bands are a sanity check on your own median, not a target handed down by the platform, and Meta publishes no official hook-rate benchmark.

Hold rate measures whether those viewers continue watching, often by dividing 15-second views by 3-second views, or by using a comparable retention step. Published hold-rate benchmarks are close to useless as a target, because they swing with video length and with which definition the publisher used, so set the floor from your own account instead. Take the median hold rate of your video ads over the last 60 days, treat that median as the survival floor, and set the scale bar about a third above it, as worked through in our guide to ad creative testing.

Don't use rigid thresholds as automatic execution rules. The ranges diagnose where to look. Your decision still depends on campaign objective, spend distribution, audience temperature, and the quality of the conversion data.

How do you read the sequence instead of the scoreboard?

The combinations expose the failure more clearly than any single score:

• High hook, low hold: The opening makes a promise the body does not fulfill. Keep the pattern interrupt, then replace the explanation, proof, or pacing.
• Low hook, high hold: The idea may be relevant, but its strongest point arrives too late. Move the proof or product tension into the first frame.
• Strong hook and hold, weak CTR: The ad earns attention without creating enough intent. Tighten the value proposition or CTA.
• Strong CTR, weak CVR: The click does not survive the landing page or offer. Check message alignment and post-click friction.

Frequency supplies the audience context. If repeat exposure rises while CTR falls and CPM rises, the creative may be fatiguing even when aggregate ROAS remains acceptable. These are the starting points we use at Selzee, set out in full in our read on ad fatigue: watch prospecting frequency at 2.5 and act at 3.0 and above, watch retargeting frequency at 4.0 and act at 6.0 and above, always read next to a falling hook rate and a rising CPM rather than on their own. Meta publishes no frequency threshold, so calibrate against your own account's normal range before any of those numbers means anything. What this means for a DTC brand: frequency is a watch signal, not a pause trigger, and it only earns a verdict when hook rate is falling beside it.

An ad with acceptable CTR but weak hold rate is a common trap. A provocative opening can generate clicks while viewers leave before the product is explained. The media result looks efficient, but the creative brief needs a new body and a more honest promise. Apply this creative diagnostics framework at the asset level, so the next decision changes the concept rather than merely shifting budget.

Reading the sequence matters because ads fail in stages. The first stage is attention. The second is comprehension and relevance. The third is intent. The fourth is conversion. When you look only at the final score, you lose the map of where the audience dropped off.

That map is what creative teams need. An editor needs to know whether the opening was too slow. A strategist needs to know whether the message overpromised. A designer needs to know whether the product benefit was clear on screen. A creator needs to know whether the proof landed late or felt unconvincing. None of those decisions comes from ROAS alone.

This is also why side by side ad review beats isolated metric review. If one concept wins hook rate while another wins CVR, the useful question is not which number is higher in isolation. The useful question is what each asset did differently at the exact step where the audience responded. That is how pattern libraries become practical rather than theoretical.

Metric Hierarchies by Campaign Objective

A dashboard can make a profitable campaign look broken, or hide a weak creative behind cheap traffic. The campaign objective decides which metric gets the final vote. Prospecting seeks qualified attention from people unfamiliar with the brand. Retargeting addresses existing intent and removes doubt. Conversion campaigns must turn that intent into profitable action.

How do the three campaign jobs change the metric hierarchy?

For prospecting, CPM describes delivery economics, not creative quality. Hook rate shows whether the opening earns attention, while hold rate tests whether the idea sustains it. New-to-brand share helps distinguish genuine acquisition from existing demand. ROAS can mislead when the campaign is introducing the product rather than capturing ready-to-buy intent.

Consideration campaigns require a different hierarchy. CTR measures the next-step response, while view-through rate and engagement rate indicate whether the message is building interest before the click. A high CTR paired with weak downstream action often points to a broad promise, sensational framing, or poor alignment with the destination. Review hook and hold together before changing the audience.

Conversion and retention campaigns should center on CPA and ROAS, supported by CVR and LTV:CAC. Creative metrics still explain why an ad wins or loses attention, but they do not outrank purchase economics when conversion is the assignment. A strong hook with weak hold can produce inexpensive clicks and little business value.

Frequency changes meaning with the objective. Repeated exposure may be acceptable in retargeting when the audience is small and intent is high. In broad prospecting, rising frequency alongside falling response signals that the concept is exhausting its audience. Put the campaign job beside the scorecard, then judge hook rate, hold rate, frequency, and business outcomes in that order.

The practical takeaway is that teams need a hierarchy before they need a dashboard. If the objective is prospecting, you are buying first contact and early interest. If the objective is consideration, you are buying movement toward evaluation. If the objective is conversion, you are buying action. The same ad can look strong or weak depending on which job it was hired to do.

Confusion often starts when people compare campaigns with different assignments as though they shared the same finish line. A retargeting campaign will naturally look more efficient on purchase metrics than broad prospecting. That does not mean the prospecting creative failed. It may mean the campaign did exactly what it was meant to do, which is to create qualified awareness that later campaigns capture.

This is where creative and performance teams need shared language. Instead of arguing about whether an ad is simply good or bad, they can ask whether it succeeded at the job attached to its objective. That framing reduces unproductive debate and makes reporting much more useful to production.

Turning Metric Verdicts Into Creative Decisions

Metrics earn their place only when they change the next brief. A scorecard that says “CTR declined” is incomplete until it shows the writer, creator, designer, or editor what to preserve, what to replace, and where the audience stopped responding.

How do you turn signals into clear verdicts?

Use four practical verdicts:

• Scale: Hook rate, hold rate, action, and conversion efficiency support more distribution. Build derivatives around the winning angle, then change one variable at a time, such as the opening, creator delivery, visual proof, or CTA.
• Iterate: One part works while another leaks. “Keep the hook, replace the CTA sequence” gives production a usable instruction. “Improve performance” does not.
• Rework: The concept has a clear diagnostic failure. Return it to strategy with the failed promise, the audience assumption, and the evidence behind the verdict.
• Kill: The ad fails to earn attention or action after a fair test. Archive the concept with a post-mortem tag so the same weak premise does not return in the next sprint.

A composite score can combine hook rate, hold rate, CTR, and conversion efficiency, but it must not conceal the component signals. Two ads with the same overall grade may require opposite decisions. One needs a stronger opening. The other needs a more credible offer.

The value of a verdict is not in the label itself. The value is in whether the label produces a specific next move. If the team says “iterate,” the brief should show exactly what survives and exactly what changes. If the team says “rework,” the brief should document which assumption broke. If the team says “kill,” the archive should preserve the lesson so the team does not revive the same weak idea under a new name.

This is where many DTC review processes fall apart. People agree on the numbers, then stop short of translating those numbers into production instructions. The scorecard becomes a report rather than an operating tool. A good verdict closes that gap.

What should a Scale brief actually say for a DTC brand?

Below is a worked example for a fictional DTC skincare brand launching a barrier repair serum. The creative verdict is Scale because the concept is doing its job and the next step is controlled expansion.

• Brand: Softwell Skin
• Product: Barrier Repair Serum
• Concept name: Redness Routine Reset
• Verdict: Scale

What the creative lead would put in the brief, field by field:

• Objective: Scale the winning concept into new variants that preserve the same core promise.
• Audience: Cold prospecting audiences interested in sensitive skin care, skin barrier support, and redness solutions.
• What to preserve: Keep the opening close-up of irritated skin, the creator line about makeup sitting badly on stressed skin, and the side-by-side before and after texture shot.
• What to change: Test new creators, new first-frame visual crops, and a shorter CTA ending. Keep the claim and product proof sequence intact.
• Core promise: This serum helps calm visible redness and supports the skin barrier without adding a complicated routine.
• Proof points to keep: Visible texture improvement, creator demonstration of application, ingredient callouts on screen, and routine simplicity.
• New variant directions: One version led by creator voiceover, one version led by on-screen text, one version led by a bathroom mirror demo, one version led by a customer-style testimonial.
• Hook instruction: Start with the feeling of skin looking irritated before the product appears. The first line should frame a familiar daily frustration, not a generic skin-care claim.
• Body instruction: Keep the explanation simple. Show the product in use within the opening moments and retain the visual proof that makes the promise believable.
• CTA instruction: Move to a shorter close that reinforces who the product is for and removes decision friction.
• Guardrails: Do not replace the core redness angle with a broad hydration claim. Do not remove the visual proof sequence.
• Success condition: The new variants should preserve the original concept's strength while revealing whether creator, opening crop, or CTA can improve efficiency.

This kind of brief is useful because it names the exact source of confidence. The team is not scaling because the ad “won.” The team is scaling because the opening, promise, and proof sequence worked together in a way worth replicating.

What should an Iterate brief actually say for a DTC brand?

Now take a fictional DTC snack brand selling a high-protein trail mix. The verdict is Iterate because one part of the ad works well, but another part breaks momentum.

• Brand: Peak Pantry
• Product: High-Protein Trail Mix
• Concept name: Desk Drawer Rescue
• Verdict: Iterate

What the creative lead would put in the brief, field by field:

• Objective: Keep the attention-winning opening and replace the weak action-driving section.
• Audience: Office workers, commuters, and people looking for convenient afternoon snacks.
• What to preserve: Keep the first-frame visual of the creator opening a desk drawer full of disappointing snack options and reacting to them.
• What to change: Replace the middle explanation and final CTA. The current copy talks too generally about healthy snacking and does not make the product feel distinct.
• Core promise: This is a convenient snack that feels satisfying rather than like a compromise.
• Problem with current version: The hook creates recognition, but the body drifts into generic category language. Viewers understand the situation but not why this product is the better choice.
• Revised proof direction: Show handful texture, ingredient close-ups, and portion convenience. Add one line that explains why it is more filling than the snacks it replaces.
• Hook instruction: Keep the drawer reveal and frustration line exactly as the starting pattern. It is doing the stopping work.
• Body instruction: Move quickly from the problem to the product. Show the texture, portability, and practical use case in sequence.
• CTA instruction: Replace the broad “try it now” ending with a more specific purchase reason tied to routine convenience.
• Guardrails: Do not change the opening concept. Do not add lifestyle filler shots before the product explanation.
• Success condition: The next cut should preserve recognition and increase intent by making the product difference obvious before the close.

An iterate brief needs to sound surgical. The team should know that they are not starting over. They are preserving the useful part of the concept and repairing the leak.

What should a Rework brief actually say for a DTC brand?

Consider a fictional DTC home brand selling a compact air purifier. The verdict is Rework because the concept rests on a strategic assumption that is not landing.

• Brand: Northroom Home
• Product: Compact Air Purifier
• Concept name: Luxury Look, Cleaner Air
• Verdict: Rework

What the creative lead would put in the brief, field by field:

• Objective: Return the concept to strategy and rebuild the angle around a clearer buyer motivation.
• Audience: Apartment renters, pet owners, and people concerned about dust and indoor air quality.
• What failed: The ad assumes people will respond to the purifier as a design object first. The audience is not showing enough interest in that framing.
• Why it failed: The opening image is attractive, but the concept delays the functional reason to care. The audience sees decor before utility.
• Audience assumption to revisit: People may value compact design, but they still need the reason for purchase to be practical and immediate.
• Reframed promise: Keep the small-space angle, but lead with the problem of stale air, dust, or pet-related irritation before introducing the product form factor.
• New proof direction: Show where the purifier fits in tight spaces, then demonstrate the practical reason it belongs there.
• Hook instruction: Replace beauty-led opening shots with a problem-led first frame that introduces the daily annoyance the product addresses.
• Body instruction: Bring product demonstration forward. Explain the fit for apartments, desks, or bedrooms through use, not abstract design language.
• CTA instruction: Close on the reason to own it, not the fact that it looks good in a room.
• Guardrails: Do not rebuild this as a purely aesthetic lifestyle ad. Do not reuse the original opening line.
• Success condition: The new concept should prove that the audience responds to the product as a useful solution before the team revisits design-first positioning.

A rework verdict should show more humility than an iterate verdict. The team is not just changing wording or pacing. It is challenging the premise underneath the ad.

What should a Kill brief actually say for a DTC brand?

Finally, take a fictional DTC apparel brand promoting wrinkle-resistant travel pants. The verdict is Kill because the concept is not generating useful attention or useful action.

• Brand: Roam Thread
• Product: Wrinkle-Resistant Travel Pants
• Concept name: Airport Compliment Test
• Verdict: Kill

What the creative lead would put in the brief, field by field:

• Objective: Retire the concept and preserve the reason it failed.
• Audience: Frequent travelers and commuters.
• What failed: The social experiment framing did not create believable product interest or purchase intent.
• Evidence summary: The opening did not earn enough qualified attention, the reactions felt staged, and the product benefit remained vague until too late in the ad.
• Why the concept should not be revised: The central device depends on a reaction format that distracts from the actual reason to buy the product.
• Lesson to archive: Audience curiosity alone is not enough when the core benefit requires direct demonstration. This category needs proof of function, not approval theater.
• Do not reuse: Public reaction hooks, compliment framing, or delayed product explanation.
• What to test instead: Direct packing demo, wrinkle comparison, all-day wear comfort, and creator-led travel use cases.
• Archive tag: Weak premise, low product clarity, low trust.
• Production instruction: Stop derivative work on this concept. Reallocate effort to function-led demonstrations.
• Success condition for replacement concepts: New ads must show the product solving a travel problem early and credibly.

A kill verdict is not wasted work if the lesson is sharp. It becomes wasted work only when the team fails to document what made the premise weak.

What does a strong weekly creative scorecard entry look like?

The weekly scorecard should not just collect metrics. It should force better language. If a field can be filled with a vague phrase, the field is not doing enough to help production.

The difference between the two columns is not polish. It is usability. A good entry gives the next person enough clarity to make something better. A weak entry forces the team to rediscover the same lesson in the next meeting.

How do you build a learning loop that the team will actually use?

The operating principle applies beyond any single system: every result should leave behind a reusable creative rule. If a hook wins only when it opens with a visual demonstration, record that condition. If a benefit claim earns attention but loses conversion because proof arrives too late, record that pattern as well. The testing library should become more specific after every cycle.

Incrementality gives that library a stronger foundation. Hosahally and colleagues, writing in the Journal of Digital & Social Media Marketing in March 2025 on measuring digital advertising in a post-cookie era, score the available measurement methods on ease of use, accuracy, validation, robustness and predictiveness, and land on the incrementality randomised control trial as the method to adopt. Worth knowing that the author list includes a measurement vendor's co-founder. Platform ROAS reports attributed efficiency. A holdout addresses the harder question: which creative caused additional demand? Our guide to multi-touch attribution covers when to trust platform credit and when to demand a control group. What this means for a DTC brand: a holdout is the only read that tells you whether a winning concept created demand or just harvested it.

A single verdict can still mislead, which is why the operating lesson matters more than any single ratio: separate media efficiency from creative signal. A cheap click can hide a weak concept, while strong hook and hold rates can show that the idea deserves a better offer, audience, or destination.

A real learning loop turns those observations into assets the team can reuse. That means naming patterns in a way that survives the meeting. Instead of writing “demo ads work,” write “demo ads work when the problem appears on screen before the product explanation.” Instead of writing “creator UGC performed best,” write “creator UGC performed best when the creator states the daily frustration in the first line and handles the product within the opening beats.” Specificity is what makes future testing smarter.

A useful learning loop also separates durable rules from temporary conditions. Some wins come from the concept itself. Others come from timing, offer strength, or audience readiness. If the team does not distinguish between those causes, the testing library becomes noisy. The purpose of the library is to help the next brief start from evidence, not from vague memory.

The weekly cadence that turns those entries into a production queue is set out in our guide to report generation for paid social, and teams that want to cut the spreadsheet work can organise the scorecard in ad reporting tools.

How Selzee runs the metric verdict into the next brief

Selzee is an AI content team with its own interface, turning customer reviews, ad comments, ad-account data, competitor ads, and the organic feed into ready-to-ship briefs, test plans, and creator matches. It carries the process beyond reporting, so a metric verdict can become a brief, a planned test, and a creator shortlist.

The part that matters is the handoff. A verdict that lives in a slide dies on Monday afternoon. A verdict that arrives as a brief with the preserved elements, the replaced elements, and the reason attached is something a writer, editor or creator can act on the same week. That is the loop our guide to creative diagnostics grades, and it is the same loop the scorecard above is built to feed.

FAQ

Which ad metric should a creative team look at first?

Start with the metric closest to attention. For video creative, that usually means asking whether the opening earned a stop before moving to click or revenue outcomes. The point is not to ignore later metrics. The point is to diagnose the sequence in the order the audience experienced it.

Why is CTR not enough to judge an ad?

CTR can tell you that people clicked, but it cannot tell you whether the ad qualified those clicks well. A curiosity-led hook can drive traffic that was never likely to convert. Without context from retention, message clarity, and post-click behavior, CTR can flatter a weak concept.

What is the difference between a hook problem and a hold problem?

A hook problem means the opening did not earn enough attention. A hold problem means the ad got initial attention but failed to maintain relevance or momentum. The fix is different in each case, which is why separating the two matters.

When should a team scale a creative concept?

Scale when the concept works as a sequence, not just as a single moment. The opening should attract the right kind of attention, the body should deliver the promise, and the ad should drive action without obvious disconnects. Scaling too early usually means scaling ambiguity.

When should a team iterate instead of rework?

Iterate when the core concept is sound and one part of the execution leaks performance. Rework when the premise itself is shaky, the audience assumption feels wrong, or the ad depends on a promise that does not hold up once the product appears.

What belongs in a useful creative verdict?

A useful verdict names the decision, the evidence behind it, what to preserve, what to change, and what production should do next. If the verdict cannot be handed to a writer, editor, or strategist as a usable instruction, it is still too vague.

If your team is still treating CTR or ROAS as the whole story, visit Selzee to turn ad data, customer feedback, and market signals into creative briefs, test plans, and creator matches. Use the diagnostic sequence to decide what to keep, what to rework, and what to ship next.