The Farm Digital
Meta Ads Audit
Performance Review
RetainersDirect · Ecommerce, DTC Orthodontic Retainers

Meta Ads Account Audit: What's Working, What's Not, and What It Should Look Like

A full read of the account structure, spend, and results for the period below, sourced directly from the two exports provided.

Reporting Window
May 1, 2026 to Aug 11, 2026 (103 days)
Data Source
Ads Manager exports (structure + performance)
Prepared By
The Farm Digital
Verdict

Degrading, and fixable. The account is not broken, but half its active budget is sitting in the wrong place. The retargeting ad set is spending almost as much as prospecting ($7,903 vs $8,145) while returning less than half the ROAS (1.23x vs 2.56x). Now that click and landing page data confirms the ads themselves are working (retargeting clicks through cheaper and more often than cold prospecting), the fix points at the site experience for returning visitors, not the creative.

Big Picture, Before Anything Else

Eight campaigns exist in this account. Two are actually running.

Before diagnosing performance, here is what the account looks like end to end. This matters because most of the account's history and creative testing lives in campaigns that are currently paused, not deleted, which changes what "starting fresh" would actually mean here.

8
Total campaigns in the account
2
Currently active and spending
6
Paused (2 Advantage+ Shopping, 4 legacy testing)
149
Distinct ads on record across the account's history
CampaignStatusObjective / Bid StrategyDaily BudgetRole
[V] CBO | DCT | Top Performers | US | CC $75ActiveOutcome Sales, Cost per result goal$70Prospecting (broad, Dynamic Creative)
598LA | Retargeting | USAActiveOutcome Sales, Highest volume or value$115Retargeting (warm custom audience)
[MM] ASC | Best Winning Ads | US CampaignPausedOutcome Sales, Highest volume or value$100Advantage+ Shopping, dormant
[V] ASC | Best Winning Ads | US CampaignPausedOutcome Sales$80 (ad set level)Advantage+ Shopping, dormant
598LA | ASC+ | SCALING | USAPausedOutcome Sales, Highest volume or value$85Legacy scaling test
598LA | CBO - Video | Testing | USAPausedOutcome Sales, Highest volume or value$85Legacy creative testing bench
598LA | CBO | TESTING | USA (Cost Cap)PausedOutcome Sales, Cost per result goal$100Legacy angle testing (12+ ad sets, angles 012 to 026)
598LA | CBO | TESTING | USAPausedOutcome Sales, Highest volume or value$85Legacy angle testing (angles 001 to 016)
Worth knowing before we touch anything

The four "598LA | CBO | TESTING" campaigns are not empty shells. They hold over 20 individually numbered, dated ad sets, each built around a single tested creative angle (savings, comparison to Invisalign, night guards, testimonials, and more). That is a real testing archive. Before writing new creative from scratch, it is worth mining what these angles were and whether any of them quietly outperformed before being paused.

How's The Performance

Account-wide numbers look fine blended. They stop looking fine once split by ad set.

$16,047.96
Total spend, 103 days
300
Purchases (tracked results)
$53.49
Blended cost per purchase
1.90x
Blended ROAS (implied revenue $30,536)

Cross-check: implied average order value from the data is $101.79, which lines up almost exactly with the $99 retainer price point advertised in the creative. That consistency is a good sign the purchase tracking itself is at least directionally reliable, more on that in the tracking section below.

Ad SetRoleSpendResultsCost / ResultROASReachFrequency
DCT Ad 2Prospecting (broad)$8,144.87216 $37.712.56x252,5521.81
Warm Audiences (ATC/WV/ENG) | USA - 7DCRetargeting$7,903.0984 $94.081.23x123,5562.92
The weakest link

Retargeting is being outperformed by cold prospecting, by a wide margin. DCT Ad 2 (broad, unrestricted, Advantage+ audience) is converting at $37.71 per purchase and 2.56x ROAS. The retargeting ad set, showing ads to people who already visited the site, added to cart, or engaged with the page, is converting at $94.08 per purchase and only 1.23x ROAS, which is barely above the point where ad spend alone breaks even before product cost and fulfillment are even factored in.

That is backwards from what a warm audience should do, and it is spending nearly as much as prospecting to get there ($7,903 vs $8,145, essentially a 50/50 split). Frequency also confirms the strain: 2.92 in retargeting vs 1.81 in prospecting, on a pool less than half the size (123,556 reach vs 252,552).

Update: now that we have click and landing page data, the break happens after the click, not before

Yomi sent through a day-by-day export with CTR, outbound CTR, and cost per landing page view included (May 1 to Aug 10, close enough to the window above to cross-check cleanly, and the CPA and ROAS in this new pull land within a few cents and a few points of the numbers already reported, so the two exports agree).

MetricDCT Ad 2 (Prospecting)Warm Audiences (Retargeting)
CTR (all)1.70%2.38%
Outbound CTR1.08%2.08%
Landing page view rate0.90%1.47%
Cost per landing page view$1.96$1.47

Retargeting wins on every single click and landing-page metric. It gets clicked more, gets people to the site cheaper, and does it at a higher rate than cold prospecting. That rules out the creative and the hook as the problem. Whatever is going wrong happens after someone from this audience already lands on the site: the offer, the landing page experience for a returning visitor, or checkout, not the ad itself.

Weekly trend adds one more distinction worth knowing before prioritizing a fix. DCT's ROAS started strong (8.36x the week of April 27) and has eroded gradually to a 1.5x to 3.2x range since, a normal fatigue curve as the campaign matures. Retargeting never had that strong start to erode from: its ROAS has bounced between 0.60x and 1.68x every single week since April 27. That points to a standing structural issue with this ad set rather than something that recently went stale.

How's The Budget

Not underfunded. Misallocated.

Combined nominal daily budget across the two active campaigns is $185 ($70 DCT + $115 retargeting). Actual average spend over the window was $155.81 a day, about 84% of that ceiling, which is normal pacing behavior for Meta's auction rather than a red flag on its own.

$185 / day
Combined nominal daily budget (both active campaigns)
$155.81 / day
Actual average spend, 84% of nominal ceiling
Before increasing budget

At a blended 1.90x ROAS with retargeting sitting at 1.23x, adding more budget right now mostly means spending more into the weaker half of the account. The recommendation is to fix the split before growing the total.

ActionWhereWhy
Hold flat, then growDCT Ad 2 (prospecting)Already returning 2.56x blended, and its best single ad returns 5.14x. This is the campaign that has earned more budget, once the internal allocation issue below is fixed.
Hold, do not addRetargetingFix the post-click experience and attribution setting first. Adding budget to a 1.23x ROAS ad set before fixing it just scales the problem.
Fund as a testOne Advantage+ Shopping campaignCurrently $0 of the $185/day nominal budget touches ASC. Worth a defined $40 to $50/day test rather than leaving Meta's primary ecommerce engine fully dormant.

Net recommendation: keep total spend roughly where it is for the first two to three weeks while the retargeting pool and attribution fix take effect, then layer in the ASC test and grow DCT from a position of confirmed efficiency rather than growing the account uniformly.

Tracking & Measurement

Signal check: what the data confirms, and what it can't tell us

Yomi sent through Events Manager screenshots and a day-by-day performance export. Most of what this audit needed to verify is now answered. A couple of items remain open.

Confirmed clean

Purchase event match quality is 9.3 out of 10, well above the 7-and-above bar for a healthy purchase event. Add payment info sits at 8.5. Conversions API is active on Purchase specifically and is reporting 200% more conversions than the Pixel alone would catch, with value, order ID, and 9 other parameters passing through.

Blended average order value ($101.79) lines up almost exactly with the advertised $99 price, and the "Results" column matches the "Purchases" column on every row with zero exceptions. Purchase-side tracking is solid, the ROAS figures in this audit can be trusted.

Confirmed as an issue

The events that build the retargeting audience are the weak spot, not Purchase. PageView sits at 5.9 match quality, View content at 5.8, both below the healthy line. Events Manager is also flagging a live "low rate of pixel events covered by Conversions API" warning, and Meta's own note on that warning says advertisers who fixed it saw lower cost per result specifically on PageView. That is likely part of why the "Web Visits" layer of the retargeting pool is thinner or noisier than it should be.

Two open items, still worth 10 minutes each

Two pixels exist on this ad account: "retainersdirect's pixel" (the one in these screenshots, actively used by the live ad sets) and a second one called "Retainers Direct Pixel." Worth confirming the second one is legacy and unused rather than quietly splitting data or double firing on the site.

Domain verification is unresolved for admin.shopify.com, flagged since June 3. Left unresolved, this can limit how many conversion events the account can prioritize and report on iOS.

Ad Set: Is The Audience Performing, Or Too Restricted

Prospecting is correctly wide open. Retargeting is wide open in the wrong way.

DCT Ad 2 (prospecting)

GeographyUS, no exclusions
Age / Gender18 to 65, all genders
Advantage+ AudienceOn (expansion_all)
Custom audiencesNone, pure broad
ExclusionsNone on the live version
ReadCorrectly broad

Warm Audiences (retargeting)

GeographyUS, no exclusions
Age / Gender18 to 65, all genders
Advantage+ AudienceOff (correct for retargeting)
Custom audiencesWeb visits 180d, ATC 180d, IG/FB engagement 365d
ExclusionsNone visible on the live pool
ReadToo wide, too stale
Audience staleness is a contributing factor, the click data points mainly to what happens after the click

The retargeting pool is not too small. It is too loosely defined for what "warm" is supposed to mean. Stacking 365-day Facebook and Instagram engagers into the same pool as 180-day add-to-cart users and 180-day site visitors dilutes genuine purchase intent with a much colder layer of people who liked a post over a year's time window, and Events Manager backs this up: match quality on the PageView and View content events that build the "web visits" layer is a weak 5.8 to 5.9, so that layer is likely thinner or noisier than Ads Manager assumes.

That said, the CTR and landing page data in the tracking section above shows this pool clicks through cheaper and more often than cold prospecting. So the pool is not failing to generate interest, it is failing to convert that interest once people land. Tightening the pool is still worth doing, but the landing page and offer experience for a returning visitor is the higher-priority fix.

Recommendation: split this into two ad sets, one for a genuinely hot window (site visitors and add-to-cart within roughly 30 to 90 days), one for the wider 180 to 365 day engagement layer at a much smaller budget, and let performance tell us whether the second one deserves to exist at all.

One targeting choice that looks off at first glance but is not: past purchasers are included in the retargeting pool rather than excluded. For a retainer brand, that is defensible. Several of the live ad angles ("still wearing that old retainer," "replace your old retainers") are clearly aimed at past buyers who may need a replacement, not first-time buyers. That is a deliberate repurchase play, not a targeting mistake.

Ad: Which One Is Actually Winning

The best ad in the account is getting the least budget

Ranked by cost per result among ads with meaningful spend in the window. "Below average" quality/engagement/conversion rankings are Meta's own relative scores against competitors reaching the same audience, shown where Meta had enough volume to generate one.

AdAd SetStatusSpendResultsCost / ResultROAS
STATIC | NEW WAY TO GET RETAINERS | 6.16DCTActive$207.5511$18.875.14x
VIDEO | 500 FOR RETAINERS?? | 4.30DCTPaused$420.5518$23.364.07x
FLEX | ADS WEEK 7.20DCTActive$575.5024$23.983.66x
STATIC | WINNER | OLD RETAINERSDCTPaused$841.5024$35.062.21x
VIDEO | OLD RETAINERS | 5.12 (DCT version)DCTPaused$2,942.8173$40.312.74x
VIDEO | APPARENTLY | 6.26 (largest active spender)DCTActive$2,838.2956$50.681.77x
VIDEO | AI GFRetargetingActive$862.4912$71.871.79x
VIDEO | OLD RETAINERS | 5.12 (retargeting version)RetargetingPaused$1,684.0124$70.171.49x
STATIC | WHY PAY DENTISTSRetargetingActive$693.218$86.651.14x
STATIC | MOF | STOP PAYING $500 | 6.17RetargetingPaused$1,186.8511$107.901.05x
STATIC | MOF | AMERICA'S SMARTEST WAY TO BUY | 6.17RetargetingPaused$537.623$179.210.78x
Budget is not following performance inside DCT

STATIC | NEW WAY TO GET RETAINERS | 6.16 is the most efficient ad in the account: $18.87 per result, 5.14x ROAS. It is getting $207.55 of the ad set's $8,144.87 in spend, about 2.5%.

VIDEO | APPARENTLY | 6.26 is getting $2,838.29 of that same budget, about 35%, at a much weaker $50.68 cost per result and 1.77x ROAS, below the ad set's own blended average.

Dynamic Creative Testing is supposed to auto-shift budget toward winners, and it has not caught up here yet. Worth a manual nudge: either increase the bid or add fresh variants around the winning angle so the algorithm has more to work with, or break it out into its own ad set so it stops competing with a creative that is currently absorbing more of the budget than its results justify.

One flag worth watching, not acting on yet: the top ad's Quality ranking shows "below average, bottom 35% of ads" even though its Conversion ranking is above average. That usually means the hook is aggressive enough to convert well now, but could be building some negative feedback over time. Keep an eye on CPM trend for this specific ad rather than pulling it while it's the best performer in the account.

Recommendation: The Right Setup To Get Better Performance

Prioritized fixes, ordered by impact against effort

1
Audit the post-click experience for returning visitors
Retargeting clicks through cheaper and more often than cold prospecting (2.38% CTR vs 1.70%, $1.47 vs $1.96 per landing page view), so the ad and the hook are not the problem. Walk the actual path a warm visitor takes: does the landing page acknowledge they have been here before, is there any cart-recovery or "still interested" messaging, does checkout work smoothly for a returning session. This is now the best-evidenced fix in this audit.
Verify by watching cost per result trend down from $94.08 and ROAS move up from 1.23x over two to three weeks once a change ships.
2
Rebuild the retargeting audience pool
Drop the 365-day Facebook and Instagram engagement layer. Keep 180-day site visitors (excluding purchasers) and 180-day add-to-cart as the core pool, and consider splitting out a tighter 30 to 90 day high-intent layer if the account has enough volume to support it. Secondary to the fix above, but still worth doing.
Verify by watching whether frequency comes down from 2.92 and cost per result improves further on top of the post-click fix.
3
Standardize the attribution setting account-wide
Move the retargeting ad set from "7-day click" only to "7-day click, 1-day view, or 1-day engaged-view," matching DCT. Right now the two ad sets are not being measured on the same basis.
Verify by confirming both ad sets show the same attribution setting going forward, then re-pull CPA for a fair comparison.
4
Rebalance ad-level spend inside DCT Ad 2
Give STATIC | NEW WAY TO GET RETAINERS | 6.16 room to absorb more of the ad set's budget, either with a manual bid nudge, fresh creative variants around the same angle, or its own ad set so it is not competing against VIDEO | APPARENTLY for delivery.
Verify by tracking whether its share of ad set spend grows and whether ad set blended CPA moves down from $37.71 toward its own $18.87.
5
Reactivate the strongest paused DCT ads
VIDEO | 500 FOR RETAINERS?? | 4.30 ($23.36 cost per result, 4.07x ROAS) and STATIC | WINNER | OLD RETAINERS ($35.06 cost per result, 2.21x ROAS) both beat the ad set's current blended average and are currently paused.
Verify by confirming their historical efficiency holds in the first one to two weeks back live before counting on them long term.
6
Close the last tracking loose ends
Fix the Conversions API coverage warning on PageView, confirm the second unused pixel ("Retainers Direct Pixel") is not quietly firing anywhere, resolve the admin.shopify.com domain verification flagged since June 3, and get the account's change log for this window to line up against the DCT fatigue curve.
Verify by confirming the Diagnostics tab in Events Manager shows zero active errors, and by checking PageView match quality moves up from 5.9.
7
Revive one Advantage+ Shopping campaign as a defined test
Start around $40 to $50 a day. Right now 100% of prospecting depends on a single DCT ad set, which is a concentration risk. ASC is Meta's current default ecommerce engine and the account has zero live exposure to it.
Verify with a clean two to three week learning window against a defined CPA target before judging it against DCT.
The Call, Per Campaign

Scale, hold, fix, or kill

[V] CBO | DCT | Top Performers | US | CC $75
2.56x blended ROAS, best ad at 5.14x. Fix the internal budget allocation and reactivate the two strong paused ads first, then this is the campaign to grow.
Fix, then scale
598LA | Retargeting | USA
1.23x ROAS, barely above ad-spend breakeven. Clicks are cheap and plentiful, so fix the post-click experience first, then the audience pool and attribution, before adding or even holding current spend levels.
Fix
[MM] ASC | Best Winning Ads & [V] ASC | Best Winning Ads
Zero spend all window. Not proven bad, just untested recently. Worth a structured revival rather than a permanent kill.
Test
The four 598LA | CBO | TESTING campaigns
Zero spend all window. No urgent action, but this is the account's creative-angle archive (20+ tested angles). Worth reviewing before writing new creative from scratch.
Hold, mine for angles