What we spent, what we learnt
and what I recommend next.
Band of Boys + The Girl Club
THE ADS.
THE NUMBERS.
WHAT’S NEXT.
Start with this
Three things to take away.
01
AW26 had fewer orders, but more from new customers.
That is the first-30-days result across the whole store. We cannot say the ads alone caused those sales.
02
Build on BoB and give TGC a fairer test.
Keep the BoB ads bringing in purchases, introduce fresh product and outfit creative, and test TGC without spreading the budget too thin.
03
Keep the current budgets while we make those improvements.
The review supports focused changes. It does not establish a full collection spend total or show that more spending would have fixed either launch.
Previous spend
There are two different
questions here.
1 · Specific campaigns
What did the campaigns
we found cost?
Use these figures to see the cost of particular launch and sale campaigns, over the dates they actually spent money.
2 · The first 30 days
What was spent across
the ad accounts?
Use this comparison to see all Meta and Google spend during the first 30 calendar days of each launch.
They overlap. Some of the same spending appears in both views, so do not add them together.
Specific campaigns · SS25
SS25: the campaigns we found.
Combined Meta spend
NZ$1,891.63
Across three campaigns,
including prelaunch activity.
Voucher giveaway3–9 September 2025$199.70
Awareness ads7–9 September 2025$212.78
Purchase campaign10–29 September 2025$1,479.15
This is a partial view of SS25 spending. Other ads may also have promoted the collection.
Specific campaigns · AW26
The launch and the later sale
are separate.
Launch awareness · Meta
NZ$249.86
28 March–1 April 2026
One campaign introducing the launch. It does not include AW26 content in other ongoing ads.
Later end-of-season sale · Meta + Google
NZ$5,283.91
9 July–6 September 2026
Four dedicated sale campaigns. It does not include sale content in campaigns that also ran other promotions.
Why the full total is still unclear
Some campaigns did
more than one job.
Some ads may have promoted several ranges or offers at once.
What we know
How much those
campaigns spent.
The campaign costs are recorded.
What we cannot confirm
How much belongs
to each collection.
The old ad and product records do not give us a reliable split.
What this means: I cannot confirm a full SS25 or AW26 spend total, or whether either collection’s advertising was profitable.
A separate comparison · first 30 calendar days
AW26 had less ad spend
in its first 30 days.
SS25
11 September–10 October 2025
Meta $3,653.72Google $2,643.75
AW26
2 April–1 May 2026
Meta $2,185.50Google $2,893.44
Same launch windows · whole-store Shopify results
Fewer orders overall.
More from new customers.
| SS25 | AW26 | |
|---|---|---|
| All orders | 252 | 223 |
| Net sales | NZ$26,359.75 | NZ$22,874.87 |
| Orders from new customers | 71 | 96 |
These results alone do not prove whether the ads were more effective.
The ongoing ads · often called “evergreen”
Keep what is bringing
people back.
01
Keep the BoB ad bringing in purchases.
Meta credited it with seven purchases from NZ$201.95 spend in the reviewed week. Keep it running while replacement wording is checked.
02
Keep searches for the brand names.
Keep the better-supported search terms, including “childrens tops”. Only the weaker keyword needs pausing.
03
Keep BoB and TGC shopping paths separate.
Each brand’s catalogue ads should show its own products and take people to the matching range.
Where I would take the ongoing ads next
Fresh creative.
A fairer test for TGC.
01
Introduce each range through products and outfits.
Use fresh product and outfit images alongside the existing catalogue ads. Keep BoB and TGC imagery, products and shopping links distinct.
02
Give TGC enough attention to learn something useful.
TGC-named Meta ads received only NZ$15.68 in the audited week. That is too little evidence to judge the range’s advertising potential.
03
Keep the testing focused.
Test a small number of ideas and use the results to guide the next creative decision. Keep the existing budgets for now.
Supporting the creative plan
A few fixes behind the scenes.
01
Keep the searches and audiences relevant.
Remove the specific weak searches, exclude recent buyers from the cart audience and make each Google campaign’s role clearer.
02
Keep the wording and shopping links accurate.
Replace unsupported urgency, protect private offers and correct old campaign labels attached to ad links.
03
Make sure the intended products can advertise.
Work through the remaining Meta product information and Shopify access issues with the brand’s technical contact.
Product information used by the ads
Google looked clear.
Meta still needs a tidy-up.
Google · checked 21 September
No products needing attention
in the NZ and AU filters.
That is useful, but it does not prove that every product page and ad link has been checked.
Meta · checked 21 September
Some products were blocked
by missing information.
There were also Shopify access warnings. Start with the products intended for advertising and fix their information in the Shopify source that supplies them.
Who handles it: the brand’s Shopify or feed contact, coordinated with Justine. Check which source owns each product before combining anything.
Your question about learning
“Learning” does not mean
we should spend more.
Meta is working out how to deliver the ads. Results can fluctuate while it learns.
~50
purchases per ad set
over seven days
The guideline checked in the audit,
after a significant change.
It is a guide,
not a guaranteed target.
It applies to one ad set, a group of ads with the same audience settings. It is not 50 orders across the website.
I would keep budgets where they are and make the focused improvements in this review.
Next steps · recommended actions
Here’s what happens next.
01Brand / billing
Check the Google payment warning.
Ask the person handling billing to resolve the warning.
02Justine
Make the focused ad changes.
Searches, recent-buyer exclusion, link labels and replacement copy.
03Brand / Shopify
Fix products intended for advertising.
Resolve missing product information and check the supplying sources.
04Justine + brand
Prepare the next creative test.
Choose product and outfit assets, with a clear role for each brand.
My recommendation
BUILD ON BOB.
TEST TGC.
KEEP IT FOCUSED.
Keep the ads bringing in purchases.
Bring in fresh product and outfit creative.
Keep budgets steady.
Plan the next Spring/Summer launch separately.
These are recommendations.
No changes from this review have been made.
The ongoing ads · what I recommend changing
Tighten where the budget goes.
01
Stop Shopping ads showing for “farmers” and “farmers nz”.
Those exact searches used NZ$60.32 in the latest seven-day review, with no conversions recorded. Keep other relevant clothing searches open.
02
Pause the keyword “kids clothing boys”.
Its recent results were weak. This is a change to that one keyword, not the whole campaign.
03
Exclude recent buyers from the cart-focused audience.
Add people who bought in the past 30 days to the exclusion list. The other audiences checked already had this setting.
The ongoing ads · Google
Give each Google campaign
a clear job.
Reach people beyond the brand names
Tighten the campaign
meant for non-brand searches.
Google’s Performance Max campaign can still show Shopping ads for the brand names. I recommend excluding both brands there, while keeping the separate brand Search ads running.
Keep the click information accurate
Fix old labels
attached to ad links.
A new-season Search ad still has labels for an older sale and another campaign type. Correct those labels. Keep the stockist ad linked to its stockist page.
The ongoing ads · wording and offers
Make sure the ads say
what we can support.
Product availability
Replace “almost gone”
with neutral wording.
The audit did not have stock evidence to support that claim. Prepare and check the replacement before changing the ad that is bringing in purchases.
Private promotions
Check what Google
can add automatically.
Review the settings that can pull offer wording and website pages into ads, so private VIP offers stay private. No leaked code was confirmed.
“Explore Band of Boys. See the styles and sizes available now.”Suggested replacement copy
Supporting detail · campaign costs
Launch campaigns we could identify.
| Campaign | Dates with spending | NZD |
|---|---|---|
| SS25 voucher giveaway | 3–9 Sep 2025 | $199.70 |
| SS25 awareness ads | 7–9 Sep 2025 | $212.78 |
| SS25 purchase campaign | 10–29 Sep 2025 | $1,479.15 |
| SS25 subtotal | The three campaigns above | $1,891.63 |
| AW26 launch awareness | 28 Mar–1 Apr 2026 | $249.86 |
The dates run from the first to the last day with spending. The SS25 campaigns include prelaunch activity. The last date above is not the end of the SS25 season.
Other ongoing ads may also have carried collection products. A full Google or combined collection total cannot be confirmed.
Supporting detail · separate from launch
The four dedicated sale campaigns.
| Campaign | Dates with spending, 2026 | Meta NZD | Google NZD |
|---|---|---|---|
| Ads for a warm audience | 9 Jul–6 Sep | $2,234.15 | N/A |
| Product catalogue retargeting | 9 Jul–6 Sep | $2,434.33 | N/A |
| Australian test | 29 Jul–18 Aug | $563.08 | N/A |
| Google sale Search ads | 12 Jul–6 Sep | N/A | $52.35 |
| Platform subtotals | $5,231.56 | $52.35 |
Combined spend: NZ$5,283.91. This excludes sale content carried in campaigns that also ran other products or promotions.
The Australian test spent money on 21 days, despite “seven day” appearing in its name. Google was set to start on 10 July; its first recorded spend was 12 July.
Supporting detail · costs known, collection split unknown
These amounts cannot be
assigned to one collection.
| Period | Campaign group | NZD |
|---|---|---|
| 22 Mar–17 Sep 2026 | Ongoing Meta campaigns | $9,905.53 |
| 22 Mar–17 Sep 2026 | Other Google campaigns | $14,836.78 |
| 22 Mar–17 Sep 2026 | Australian test (“proof test”) | $244.66 |
| Jan–Feb 2026 | Meta campaigns reaching new people | $3,498.87 |
| Jan–Feb 2026 | Three campaigns with sale-related names | $1,129.28 |
These groups may include collections, sales and everyday products. I cannot confirm a reliable split by collection. Do not add them to the 30-day totals.
A January sale did run. Conflicting dates and missing product history prevent assigning all of that activity to SS25.
Supporting detail · all account spending
The same length of time
for both launches.
| First 30 calendar days | Meta NZD | Google NZD | Total NZD |
|---|---|---|---|
| SS25 · 11 Sep–10 Oct 2025 | $3,653.72 | $2,643.75 | $6,297.47 |
| AW26 · 2 Apr–1 May 2026 | $2,185.50 | $2,893.44 | $5,078.94 |
What is included?
All advertising in each account during those dates, including ongoing ads alongside the launch.
What should we keep in mind?
Meta uses Pacific account dates. Google uses NZ account dates. Shopify uses the store’s day boundaries. The calendar windows match in length, but not exactly hour for hour.
Some spending also appears in the campaign breakdowns. Do not add these views together.
Supporting detail · whole-store results
The store results in those windows.
| Shopify measure | SS25 | AW26 |
|---|---|---|
| All orders | 252 | 223 |
| Net sales NZD | $26,359.75 | $22,874.87 |
| Total sales NZD | $32,585.35 | $27,611.77 |
| Orders from new customers | 71 | 96 |
| Orders from returning customers | 181 | 127 |
SS25: 11 September–10 October 2025. AW26: 2 April–1 May 2026.
Net sales and total sales differ, so compare the same row for both launches. These whole-store results include sales from other sources.
August’s tracking migration cannot verify earlier ad attribution. Do not count Meta and Google purchase credits as extra Shopify orders.
Supporting detail · Google searches
Why I recommend these two changes.
Block only the exact “farmers” and “farmers nz” searches.
18 August–16 September: NZ$191.67 spent, one Google-credited conversion and NZ$53.96 in conversion value.
10–16 September: NZ$60.32 spent, with no conversions recorded. Blocking the exact searches keeps other relevant clothing searches available.
Pause only the broad keyword “kids clothing boys”.
In the 30-day review: NZ$80.72 spent, 0.61 Google-credited conversions and NZ$84.47 in conversion value.
A fraction is advertising credit shown by Google, not part of a real order. Keep the better-supported brand terms and “childrens tops”.
Supporting detail · Meta ads
Keep the useful ads.
Fix the specific gaps.
Exclude recent buyers from the live cart-focused audience.
The 30-day buyer exclusion was missing there. Other checked audiences and the separate drafts already had it. Meta’s audience suggestions can reach beyond cart visitors; they are not a strict limit.
Check new wording before replacing the current BoB ad.
Meta credited the ad with seven purchases on NZ$201.95 spend in the dated review. Its “almost gone” wording was not supported by stock evidence. Check every displayed version of the replacement.
Keep each brand’s products and links matched.
Keep the existing BoB and TGC catalogue selections separate and hide genuinely sold-out sizes. Product-matched catalogue ads are different from a manually built carousel.
Supporting detail · Google controls
The settings behind the recommendations.
Give the non-brand campaign a clear role.
The Performance Max brand exclusion still allows Shopping ads for branded searches. I recommend strict non-brand settings covering both brands, with the separate brand Search ads kept running.
Correct old labels on the new-season Search ad.
Its link still includes “PMax” and “EOSS25” labels. Check labels inherited from account or campaign settings before editing. Keep the stockist ad on its legitimate stockist page.
Keep private offers out of public ads.
Review optimised promotions, text customisation and URL expansion. These settings can add offer text or choose website pages automatically. No private-code leak was proven.
Review where Google ads can appear.
Review the content suitability settings. This applies across the Google account, so its wider effect needs checking before changes are made.
Supporting detail · scope and remaining work
What this review can confirm.
Checked in this review
The named campaign costs and dates.
The separate 30-day account and store figures.
The specific ad findings and recommended actions.
The August purchase-tracking migration and follow-up are complete.
Still needs further checking
The full cost of each collection.
Every product source, product page and final ad link.
The Meta products with missing information or access warnings.
The Google payment warning. Ads still spent NZ$46.31 on 19 September; a lost-sales amount is not established.
The historical Google export was unavailable. The checked costs remain usable, but cannot give a full product-by-product split.
Supporting detail · the everyday ads
Why TGC needs a clearer test.
Very little spend reached the TGC-named ads.
The 9–15 September Meta review recorded NZ$15.68 for TGC-named ads out of NZ$286.11 spent in the reviewed window, using Pacific account dates.
This is a dated finding, not a current spend figure. It is insufficient evidence to judge TGC’s advertising potential in either direction.
The recommended creative direction
Introduce each range with fresh product and outfit creative, while keeping the existing separate BoB and TGC catalogue ads. Give each audience a clear job and test a small number of ideas.
Test the creative within the current budgets and review the results before deciding what to expand. These tests have not yet been launched.
Supporting detail · platform guidance checked in the audit
Learning is a delivery stage,
not a spending target.
Meta
The guideline checked was around 50 purchases credited to one ad set within seven days after its last significant edit. An ad set is a group of ads with the same audience settings.
It is not 50 orders across the website or a guaranteed pass mark. Significant edits can restart learning, although not every budget increase does. Waiting alone may not solve low purchase volume.
Its Smart Bidding learning period depends on conversion volume, the time from click to conversion and the bidding strategy. Meta’s seven-day guideline does not apply.
Recommendation: keep the current budgets and make the focused improvements in this review.
Supporting detail · sources
Where the information came from.
Campaign and store figures
Meta daily and full-period reports, reconciled on 19 September.
Retained Google and Shopify comparison records. Google dates checked on 20 September.
Historical ad and creative checks, 20–21 September.
Ad and product checks
Google and Meta audit, 17 September. TGC spend and creative direction: Sarab Ads Report, 18 September, pages 2–3. Search links and NZ/AU product checks, 21 September.
Completed August tracking sign-off and September account-cleanup review.
Learning guidance
Google: Smart Bidding learning ↗
A note about the numbers
“Credited” or “attributed” means the advertising platform counted the purchase or conversion towards an ad. It is not an additional Shopify order.
All monetary figures are NZD. This is a dated review, not a live account view.