Pettable: $2.12M Wasted Google Ads Spend Identified, 14% Better CAC
BlueAlpha's Bayesian MMM and geo-based tests identified $2.12M in annualized wasted spend across Pettable's Google, TikTok, and Meta campaigns. Blended CAC fell 14% over 9 months.
Incrementality Proven
Wasted Spend Eliminated
Spend Scaled Profitably
"We always had a hunch we were overspending on paid search, but couldn't pinpoint exactly where or by how much. BlueAlpha didn't just reveal the waste 3 they gave us the exact actions to take at the campaign level, week after week. Within 25 days, we cut Google Search spend by 10% without losing a single sale. That's when we knew this wasn't another measurement tool 3 it was our new growth operating system."
"We always had a hunch we were overspending on paid search, but couldn't pinpoint exactly where or by how much. BlueAlpha didn't just reveal the waste 3 they gave us the exact actions to take at the campaign level, week after week. Within 25 days, we cut Google Search spend by 10% without losing a single sale. That's when we knew this wasn't another measurement tool 3 it was our new growth operating system."
Jack Trent Co-CEO at Pettable
BlueAlpha deployed a Bayesian MMM with weekly refresh and geo-based incrementality tests across Pettable's paid media portfolio. Over nine months, the system identified $2.12M in annualized wasted spend on non-incremental Google Search campaigns, validated Meta at 3.4x ROI, proved TikTok incrementality in targeted audience tests, eliminated Reddit after proving zero incrementality, and reduced blended CAC by 14% with no revenue loss. The first actionable optimization shipped 25 days after integration.
Pettable: DTC Telehealth With Rising Blended CAC
Pettable (part of Receptive Inc.) is a DTC telehealth company whose primary products include pet wellness certifications and telehealth consultations. The growth team ran paid media across TikTok, Google Search (three separate accounts, including two third-party verification accounts), and Meta, with growth profit as their primary KPI.
When Google Search spend crept past break-even and blended CAC climbed roughly 20%, the team faced a specific measurement gap. As Andrew, Pettable's Performance Marketer, described the core tension: "The CAC for Meta looks strong, but when we push spend to Meta, that's not reflected in the results that we see coming in. And we don't know how much of this is seasonality, and how different this would look if we were in peak season, pushing spend."
Gabriel, who joined Pettable's team in a strategic and data role, put the need directly: "Having a clear picture of what the incremental CAC per channel is."
Platform CPAs Looked Healthy but Net Income Didn't Follow
Pettable's platform dashboards showed healthy CPAs across channels, but the correlation to net income was not there. Gabriel described the gap: "The correlation to net income just isn't there. Like it feels like Meta is way over reporting." The team's own internal sensitivity tests (turning channels on and off and watching net income) consistently showed worse results than what the platform CAC numbers would predict.
The problem was specific to Pettable's business: demand is seasonal, with clear high and low periods. During low-demand windows, rising CACs created urgency to identify which channels were genuinely driving full fair purchases (Pettable's term for a new, full-price customer completing their first transaction) vs. claiming credit for demand that would have arrived organically through direct search.
An additional complication: Pettable ran three separate Google Ads accounts, and the team suspected overlap between them. As Jack Trent, Pettable's Co-CEO, asked BlueAlpha: "Can we separate the Google Ads accounts to see if one is incremental but the other isn't, maybe?"
BlueAlpha's Approach
BlueAlpha executed a four-phase methodology with weekly optimization cycles.
Phase 1: Rapid integration and baseline MMM (Week 1)
BlueAlpha ingested historical spend and full fair purchase data across all channels and built calibrated response curves using a Bayesian hierarchical MMM with channel-specific adstock and saturation parameters. The model accounted for Pettable's seasonal demand patterns and appointment availability constraints (Pettable's service depends on provider supply, which varies by state). The initial read identified diminishing returns on Google Search at current spend levels.
Phase 2: Google Search optimization (Days 25-98)
With the MMM identifying Google Search as the highest-waste channel, BlueAlpha ran campaign-level analysis to isolate which accounts were non-incremental. The first precision cut (~10% of Google Search spend) shipped on Day 25 with zero revenue loss, capturing approximately $500K in annualized savings. Subsequent analysis identified two additional Google Ads accounts that were entirely non-incremental, producing roughly $750K and $900K in additional annualized savings when shut down.
Jack Trent described the shift: "We always had a hunch we were overspending on paid search, but couldn't pinpoint exactly where or by how much. BlueAlpha didn't just reveal the waste; they gave us the exact actions to take at the campaign level, week after week. Within 25 days, we cut Google Search spend by 10% without losing a single sale."
Andrew framed the same result from an operational perspective: "Savings achieved by being given the confidence that one of our Pettable TPV accounts could be wound down and provided a lot of savings."
Phase 3: Channel validation via geo-based incrementality tests (Days 98-240)
BlueAlpha ran geo-based holdout tests on Meta and TikTok to validate the MMM's contribution reads. Meta returned a 3.4x ROI with positive incremental lift, giving Pettable the data to scale Meta budget by roughly 40%. A TikTok audience test generated approximately 900 incremental conversions. Reddit returned zero incrementality; the campaign was terminated immediately and the budget was redeployed to validated channels.
Phase 4: Continuous monitoring
Weekly model refreshes tracked the impact of each optimization. The system detected when CPMs shifted, when creative fatigued, and when channel contribution changed, producing campaign-level recommendations each week.
Cutting Google Ads Spend Without Losing Revenue: $2.12M
Channel | Platform View | Incremental Reality (Measured) | Action Taken |
|---|---|---|---|
Google Search (TPV Account 1) | Healthy CPA, positive ROAS | Entirely non-incremental at that spend level | Account shut down (~$900K annualized savings) |
Google Search (TPV Account 2) | Positive metrics | Non-incremental | Account shut down (~$750K annualized savings) |
Google Search (core) | Strong brand query performance | ~10% of spend non-incremental | Precision cut, ~$500K annualized savings |
Meta | Healthy reported metrics | 3.4x ROI validated via geo holdout | Budget scaled ~40% |
TikTok | Strong CPA | ~900 incremental conversions from audience test | Audience-level scaling |
Positive platform metrics | Zero incrementality | Campaign terminated, budget redeployed |
14% CAC Reduction, Reddit Eliminated, Meta Scaled
$2.12M in annualized savings identified and actioned across non-incremental Google Search campaigns. The savings came from three distinct optimizations: a precision keyword/spend reallocation (~$500K) and two full TPV account shutdowns (~$750K and ~$900K).
14% blended CAC reduction sustained over nine months. The improvement came from removing non-incremental spend (which inflated blended CAC without contributing incremental full fair purchases) and reallocating to channels with proven lift.
Incremental revenue generated from scaling validated channels. Meta's 3.4x ROI gave Pettable the data to increase Meta budgets with conviction.
Reddit eliminated. The geo-based incrementality test showed zero incremental lift from Reddit campaigns despite positive platform-reported metrics. The budget was redeployed to Meta and TikTok where incrementality had been proven.
25 days from integration to first optimization. BlueAlpha's investment achieved ROI within the first 14 days.
As Jack summarized: "By learning where our spend is most incremental we were able to shift budget away from Google to paid social and that has improved our CAC."
Testing Extended to 17-State Meta Holdout and CTV
Pettable expanded the measurement system to additional channels and test types. A Meta holdout test was launched across 17 states to validate national scaling potential. A Google AI Max test ran in parallel with geo-lift validation. The testing roadmap extended to CTV as an upper-funnel channel, while the weekly MMM continued tracking marginal efficiency across the full portfolio.
Platform metrics look healthy but your net income doesn't follow? Book a 30-minute strategy call to see how geo-based incrementality tests can separate real channel contribution from platform-reported noise.
"We always had a hunch we were overspending on paid search, but couldn't pinpoint exactly where or by how much. BlueAlpha didn't just reveal the waste 3 they gave us the exact actions to take at the campaign level, week after week. Within 25 days, we cut Google Search spend by 10% without losing a single sale. That's when we knew this wasn't another measurement tool 3 it was our new growth operating system."
Jack Trent Co-CEO at Pettable
BlueAlpha deployed a Bayesian MMM with weekly refresh and geo-based incrementality tests across Pettable's paid media portfolio. Over nine months, the system identified $2.12M in annualized wasted spend on non-incremental Google Search campaigns, validated Meta at 3.4x ROI, proved TikTok incrementality in targeted audience tests, eliminated Reddit after proving zero incrementality, and reduced blended CAC by 14% with no revenue loss. The first actionable optimization shipped 25 days after integration.
Pettable: DTC Telehealth With Rising Blended CAC
Pettable (part of Receptive Inc.) is a DTC telehealth company whose primary products include pet wellness certifications and telehealth consultations. The growth team ran paid media across TikTok, Google Search (three separate accounts, including two third-party verification accounts), and Meta, with growth profit as their primary KPI.
When Google Search spend crept past break-even and blended CAC climbed roughly 20%, the team faced a specific measurement gap. As Andrew, Pettable's Performance Marketer, described the core tension: "The CAC for Meta looks strong, but when we push spend to Meta, that's not reflected in the results that we see coming in. And we don't know how much of this is seasonality, and how different this would look if we were in peak season, pushing spend."
Gabriel, who joined Pettable's team in a strategic and data role, put the need directly: "Having a clear picture of what the incremental CAC per channel is."
Platform CPAs Looked Healthy but Net Income Didn't Follow
Pettable's platform dashboards showed healthy CPAs across channels, but the correlation to net income was not there. Gabriel described the gap: "The correlation to net income just isn't there. Like it feels like Meta is way over reporting." The team's own internal sensitivity tests (turning channels on and off and watching net income) consistently showed worse results than what the platform CAC numbers would predict.
The problem was specific to Pettable's business: demand is seasonal, with clear high and low periods. During low-demand windows, rising CACs created urgency to identify which channels were genuinely driving full fair purchases (Pettable's term for a new, full-price customer completing their first transaction) vs. claiming credit for demand that would have arrived organically through direct search.
An additional complication: Pettable ran three separate Google Ads accounts, and the team suspected overlap between them. As Jack Trent, Pettable's Co-CEO, asked BlueAlpha: "Can we separate the Google Ads accounts to see if one is incremental but the other isn't, maybe?"
BlueAlpha's Approach
BlueAlpha executed a four-phase methodology with weekly optimization cycles.
Phase 1: Rapid integration and baseline MMM (Week 1)
BlueAlpha ingested historical spend and full fair purchase data across all channels and built calibrated response curves using a Bayesian hierarchical MMM with channel-specific adstock and saturation parameters. The model accounted for Pettable's seasonal demand patterns and appointment availability constraints (Pettable's service depends on provider supply, which varies by state). The initial read identified diminishing returns on Google Search at current spend levels.
Phase 2: Google Search optimization (Days 25-98)
With the MMM identifying Google Search as the highest-waste channel, BlueAlpha ran campaign-level analysis to isolate which accounts were non-incremental. The first precision cut (~10% of Google Search spend) shipped on Day 25 with zero revenue loss, capturing approximately $500K in annualized savings. Subsequent analysis identified two additional Google Ads accounts that were entirely non-incremental, producing roughly $750K and $900K in additional annualized savings when shut down.
Jack Trent described the shift: "We always had a hunch we were overspending on paid search, but couldn't pinpoint exactly where or by how much. BlueAlpha didn't just reveal the waste; they gave us the exact actions to take at the campaign level, week after week. Within 25 days, we cut Google Search spend by 10% without losing a single sale."
Andrew framed the same result from an operational perspective: "Savings achieved by being given the confidence that one of our Pettable TPV accounts could be wound down and provided a lot of savings."
Phase 3: Channel validation via geo-based incrementality tests (Days 98-240)
BlueAlpha ran geo-based holdout tests on Meta and TikTok to validate the MMM's contribution reads. Meta returned a 3.4x ROI with positive incremental lift, giving Pettable the data to scale Meta budget by roughly 40%. A TikTok audience test generated approximately 900 incremental conversions. Reddit returned zero incrementality; the campaign was terminated immediately and the budget was redeployed to validated channels.
Phase 4: Continuous monitoring
Weekly model refreshes tracked the impact of each optimization. The system detected when CPMs shifted, when creative fatigued, and when channel contribution changed, producing campaign-level recommendations each week.
Cutting Google Ads Spend Without Losing Revenue: $2.12M
Channel | Platform View | Incremental Reality (Measured) | Action Taken |
|---|---|---|---|
Google Search (TPV Account 1) | Healthy CPA, positive ROAS | Entirely non-incremental at that spend level | Account shut down (~$900K annualized savings) |
Google Search (TPV Account 2) | Positive metrics | Non-incremental | Account shut down (~$750K annualized savings) |
Google Search (core) | Strong brand query performance | ~10% of spend non-incremental | Precision cut, ~$500K annualized savings |
Meta | Healthy reported metrics | 3.4x ROI validated via geo holdout | Budget scaled ~40% |
TikTok | Strong CPA | ~900 incremental conversions from audience test | Audience-level scaling |
Positive platform metrics | Zero incrementality | Campaign terminated, budget redeployed |
14% CAC Reduction, Reddit Eliminated, Meta Scaled
$2.12M in annualized savings identified and actioned across non-incremental Google Search campaigns. The savings came from three distinct optimizations: a precision keyword/spend reallocation (~$500K) and two full TPV account shutdowns (~$750K and ~$900K).
14% blended CAC reduction sustained over nine months. The improvement came from removing non-incremental spend (which inflated blended CAC without contributing incremental full fair purchases) and reallocating to channels with proven lift.
Incremental revenue generated from scaling validated channels. Meta's 3.4x ROI gave Pettable the data to increase Meta budgets with conviction.
Reddit eliminated. The geo-based incrementality test showed zero incremental lift from Reddit campaigns despite positive platform-reported metrics. The budget was redeployed to Meta and TikTok where incrementality had been proven.
25 days from integration to first optimization. BlueAlpha's investment achieved ROI within the first 14 days.
As Jack summarized: "By learning where our spend is most incremental we were able to shift budget away from Google to paid social and that has improved our CAC."
Testing Extended to 17-State Meta Holdout and CTV
Pettable expanded the measurement system to additional channels and test types. A Meta holdout test was launched across 17 states to validate national scaling potential. A Google AI Max test ran in parallel with geo-lift validation. The testing roadmap extended to CTV as an upper-funnel channel, while the weekly MMM continued tracking marginal efficiency across the full portfolio.
Platform metrics look healthy but your net income doesn't follow? Book a 30-minute strategy call to see how geo-based incrementality tests can separate real channel contribution from platform-reported noise.
