Paid Social case, Tech Company

The ads are fine.The click goes nowhere.

A read of Q1 2025 on Meta, and a plan for the $1,000,000 Q3 launch.
Prepared for DEPT by Martin Borovski.
Download the presentation and workbook

Present with → ← or scroll

The short version

Six countries ran in Q1. Two of them cost more than a sale is worth.

spent on Meta in Q1 2025
purchases, of which 20% are incremental
incremental CPA across the account
2 of 6
countries above the $1,000 ceiling: Switzerland and Germany

Part 1. What happened

  1. The Q1 report table, by country
  2. Who is above $1,000 incremental CPA
  3. The KPI that explains it
  4. What to fix before the launch

Part 2. What to do in Q3

  1. Channels, campaign types and goals
  2. Where the $1,000,000 goes
  3. How the Meta account is built
  4. Three experiments, and eight open questions

Part 1. The report table

One workbook. Every number traceable.

The client export had 55 rows, 31 of them empty, and a February 29th that never happened. This is the cleaned version. The table below is the same one that is in the Excel file.

20%
$1,000
Download the Excel file
Tech_Company_Q1_2025_Report.xlsx

Before trusting the table

Seven places this analysis could quietly go wrong.

Each of these changes the answer. This is what I checked, and what I did about it.

The trapWhat happens if you miss itWhat I did
Q1 means January to MarchThe file runs to May 27. Use all of it and four more countries appear above the ceiling.Report is filtered to Q1. April and May sit on a separate tab.
Incremental CPA goes up, not downMultiplying CPA by 20% makes every country look five times better than it is.Spend / (purchases x 20%). A $1,000 ceiling is a $200 platform CPA.
Totals are not averagesAveraging six country CPAs gives $419. The real account CPA is $62.Every total is recomputed from summed spend, clicks and purchases.
CTR is already a percentageFormat the source column as a percent and CTR reads 16% instead of 0.16%.Recalculated from link clicks / impressions.
Frequency is not in the fileIt has to be derived, and monthly reach cannot simply be added up.Impressions / reach, reported as average monthly frequency with a note.
Dirty rowsA February 29th, 31 empty rows, blank purchases, revenue on $0 spend.Fixed or removed, each one logged on the Data Notes tab.
The obvious suspect is wrongCPC and frequency look like the story. They explain almost none of the gap.Walked the funnel step by step. See the next two sections.

Who is above the ceiling

Switzerland costs twice the limit. Germany costs nine times.

Incremental CPA by country, Q1 2025. Spend divided by the purchases that would not have happened without the ads.

Move the incrementality slider in the report section and this chart follows. At 40% incrementality Switzerland drops under the ceiling; Germany never does. The whole verdict rests on one assumption, which is why testing it is experiment number one.

What stands out

Same auction. Same ads. A very different website visit.

Each KPI for Switzerland and Germany next to the four efficient markets pooled. Everything before the click looks normal. Everything after it collapses.

Efficient markets (NZ, UK, FR, BE)SwitzerlandGermany

Before blaming the market

Three things I would say out loud to the client.

The comparison is not like for like

  • New Zealand and the UK took 92% of Q1 spend, most of it in January at about $50 CPA.
  • Switzerland and Germany ran almost entirely in March, on about $20,000 each.
  • By March, New Zealand's own CPA had risen to $87. Against that fairer benchmark Switzerland is still 4 times worse and Germany 20 times.

The samples are small

  • Germany: 10 purchases. Switzerland: 52.
  • Neither market came close to the 50 conversions a week Meta needs to finish learning.
  • The direction is safe to act on. The exact multiple is not.

Some numbers need a second look

  • New Zealand reached 12.0 million accounts in January. The country has about 5.3 million people.
  • A market that size absorbing $488,000 at a frequency of 8 is saturated or mislabelled. Its Q3 budget stays at run-rate until this is explained.
  • April and May tests in Austria, the US, Italy and Australia show the same low conversion rate. It is a pattern, not two unlucky countries.

Before Q3

Fix the visit, then buy more of them.

The greatest issue is click to purchase conversion. More budget, new audiences or new creative will not move it. In order:

Prove the purchases are being counted

Reconcile Meta purchases with shop orders for Germany and Switzerland in March. Check pixel and Conversions API, consent banner behaviour and event match quality. In the same audit, confirm New Zealand's location targeting and country breakdown.If the shop saw orders Meta did not, this is a measurement problem and the fix is a week of work.

Localise the landing page

Language (German, French and Italian for Switzerland), currency in CHF and EUR, delivery time, duties and returns stated up front.

Repair the checkout

Local payment methods, guest checkout, full shipping cost before the last step. Hypotheses to confirm with site analytics: TWINT in Switzerland, PayPal and pay-by-invoice in Germany.

Give the algorithm enough signal

One consolidated German-speaking campaign instead of thin country tests. Optimise for add to cart until 50 purchases a week is realistic, then switch to purchase.10 purchases in a quarter cannot train a purchase-optimised campaign.

Gate the budget

Hold both markets at test level with a $200 cost cap, which equals $1,000 incremental. Release launch budget only when the conversion rate holds for two weeks.

Part 2. Channels

Meta carries the launch. Local platforms open Asia.

A considered purchase at around $730 a basket needs reach first and a conversion engine second. Each channel has one job.

Meta

The conversion engine in every market.

Why
The only channel with proof: 11,161 purchases and a trained pixel. Instagram is strong in all three new markets.
Campaign types
Advantage+ sales for the launch and for existing products, video and waitlist sign-ups for the tease, catalog retargeting.
Optimised for
Purchase. Add to cart in markets below 50 purchases a week.

TikTok

Demand creation around launch week.

Why
The cheapest way to show a new device in use, through creators. Feeds retargeting pools everywhere else.
Campaign types
Smart+ web campaigns with an automatic campaign budget (TikTok's CBO), running Spark Ads from creators. Reach and video views only for the July tease.
Optimised for
Purchase in Smart+; add to cart where volume is low. Spark and non-Spark ads in separate Smart+ campaigns, as Smart+ cannot mix them.

Reddit

Where a tech device gets researched before it gets bought.

Why
Launch threads, reviews and comparisons live in tech communities. Used in the core markets only, led by the UK.
Campaign types
Community and interest targeted conversation ads at launch, then traffic and conversions with the Reddit pixel.
Optimised for
Landing page views, then purchase. Judged on assisted and incremental sales, not last click.

LINE, Kakao and Xassumption

Local presence in Japan, Taiwan and South Korea, at test size.

Why
LINE is the daily app in Japan and Taiwan, KakaoTalk in South Korea, and X is unusually large in Japan for tech news.
Campaign types
Reach and traffic to the localised launch page only.
Honest note
I have not run these three platforms myself. I would set them up with DEPT's local teams and scale only on evidence.

Not in the plan: LinkedIn (wrong mindset for a consumer device), Snapchat and Pinterest (small in the three new markets). Shares are of the full $1,000,000; the remaining 12% is the unassigned reserve.

The rough media plan

$1,000,000, and a reason for every share.

Keep the products that already sell running, launch the new device where it can be measured, enter Asia with enough to learn, keep weak markets on a short lead and hold a reserve for September.

Existing products stay on

28% of the budget, in proven markets only.

  • Core markets returned 12.5x ROAS at a $59 CPA in Q1. Switching that off for a launch would cut revenue the client already has.
  • Keeps the pixel trained and the account's results stable while the new product goes through learning.
  • Existing buyers become the first audience for the new device.

July is not awareness only

$200k in July: $120k existing products, $80k new product.

  • Most of July runs conversion campaigns for existing products, so blended ROAS holds before launch.
  • The new product tease optimises for waitlist sign-ups (or pre-orders, if the client allows), so July has a cost per sign-up, not only views.
  • Sign-ups and video viewers are the launch-day retargeting pool.

New Zealand is held, not scaled

Part of the core group, at run-rate until explained.

  • January reach of 12.0 million against a population of about 5.3 million needs an answer before more money goes in.
  • Check location targeting (living in versus recently in), the country breakdown in Ads Manager and how the export was labelled.
  • If spend is leaking outside New Zealand, that share moves to the UK, France and Belgium.

All shares are planning assumptions and sit in editable cells on the Q3 Media Plan tab of the workbook. Launch is assumed in early August. Australia moved from growth to gated: its April test ran at $1,943 incremental CPA.

Meta campaign build

Few campaigns, broad audiences, many ads.

The structure exists to concentrate signal. Every split costs learning, so each one needs a reason: language, market maturity or funnel stage.

CampaignObjective and optimisationBudget and bidMarkets
Always-on salesAdvantage+ sales and catalog ads for existing products. Purchase.Campaign budget (CBO), cost cap $200Core markets and Poland, all quarter
Launch teaseWaitlist sign-ups (or pre-orders), plus reach and ThruPlay.Campaign budget, frequency cap 2 a weekAll launch markets, grouped by language
Launch salesAdvantage+ sales for the new product. Purchase.Campaign budget (CBO). Highest volume first, cost cap $200 after learningUK, New Zealand, France, Belgium and Poland
Asia salesSales. Add to cart, switching to purchase at 50 a week.One campaign per country, own budget so none is starvedJapan, South Korea, Taiwan
Gated marketsSales. Add to cart until volume allows purchase.Cost cap $200 from day one, budget gated on conversion rateGermany, Switzerland, Austria, Italy, US, Australia
RetargetingAdvantage+ catalog ads. Purchase.About 10% of Meta spend, cappedSign-ups, video viewers, visitors, cart abandoners

Campaign level

  • CBO on wherever markets are comparable, so budget flows to the cheapest purchase.
  • CBO off across Asia: separate campaigns, or Japan would take everything.
  • Cost cap at $200, the platform CPA that equals a $1,000 incremental CPA.
  • 7-day click, 1-day view attribution, reported next to incremental results.

Ad set level

  • Broad with Advantage+ audience. Customer lists and video viewers as suggestions, buyers excluded.
  • Advantage+ placements. Reels and Stories get purpose-made 9:16 assets.
  • DLO on in multi-language markets: Switzerland and Belgium serve each person their own language from one ad set.
  • One or two ad sets per campaign. Budget sized for 50 conversions a week.

Ad level

  • 6 to 8 ads per ad set at launch, each a different idea, not a different colour.
  • Formats: Reels demo video, feature carousel, hero static with price, creator partnership ads, catalog.
  • Local language and local proof in every market. No subtitled English in Asia.
  • Refresh every 2 to 3 weeks, or when frequency passes 3 and CTR falls by a fifth.

The extra 10%

Three experiments, in the order they change decisions.

$100,000 of test budget. Each one has a decision attached before it starts.

1. Is incrementality really 20%?

Hypothesis
A flat 20% hides big differences: lower in saturated New Zealand, higher in new markets.
Design
Meta conversion lift with a holdout, in one core market and in Japan, over four weeks from launch.
Metric
Incremental purchases and incremental CPA per market.
Decision
Reset the $200 cost cap per market and re-rank the whole budget.

2. Does localisation fix the click?

Hypothesis
A local page and checkout lifts German and Swiss conversion from under 0.7% to above 1.3%.
Design
A/B test in Meta: same ads, same audience, global page against localised page.
Metric
Click to purchase rate, then incremental CPA.
Decision
Pass: release the gated budget and repeat the template in Asia. Fail: stop spending there.

3. Creators or brand in Asia?

Hypothesis
Local creator partnership ads beat adapted global brand assets on cost per add to cart.
Design
Split test on Reels and TikTok in Japan and South Korea, equal budgets, three weeks.
Metric
Cost per add to cart and hold rate at 3 seconds.
Decision
Sets the creative production mix for Q4 scaling.

Next in the queue if budget remains: cost cap against highest volume bidding, and TikTok's incremental reach on top of Meta.

What the brief does not say

Eight questions I would ask before spending a dollar.

The plan works with assumptions, as the brief allows. These are the ones that would change it.

About the numbers

  • Why is the ceiling $1,000 when the average order is about $734? If it reflects lifetime value, I would optimise for value, not volume.
  • How was 20% measured, and where? A rate measured in saturated New Zealand will understate new markets.
  • Is New Zealand really New Zealand? 12 million reached and 71% of spend in a country of 5.3 million.
  • What happened in January? 65% of the year's spend at half the later CPA looks like a promotion or a previous launch.

About the launch

  • When exactly is launch day, and what does the device cost? The phasing and the cost cap both depend on it.
  • Does the $1,000,000 also fund always-on sales of existing products? I assumed yes, with 28% kept on them.
  • Are site, checkout, payment and delivery ready in Japanese, Korean and Traditional Chinese? Q1 shows what happens when they are not.
  • Is the $1,000,000 for all markets or only the new ones? I assumed all markets, with the 10% test budget on top.
  • What creative exists, and can we work with local creators? The ad-level plan needs 6 to 8 distinct ideas per market.

The link between both parts: Switzerland and Germany failed after the click because the visit was not local. Japan, South Korea and Taiwan are the same risk at seven times the budget.

One more thing

A short film about a spreadsheet.

Forty-eight seconds. No sheep were retargeted.

Thank you. Questions are the best part.

Martin Borovski

The presentation covers the same analysis in a formal 19-slide format, with speaker notes.

A/B test: should Martin join DEPT?No results yet. The test starts at the report table.