Head-to-Head · 14 Months · Same Product · Same Industry
Same company. Same product. Same 14-month window. One is managed by the corporate template, the other by an outside agency. The data tells a clear story about what happens when you get the fundamentals right versus wrong.
Account Overview
Both accounts ran for the same 14-month period selling the same product in comparable metro areas. The differences are entirely structural — how the accounts are configured, not what they sell.
Boston wins on every efficiency metric
3.8x lower CPL · 2.6x more leads/month · 30% less spend
VB
$464
BOS
$123
VB
30
BOS
78
VB
$13,852
BOS
$9,673
VB
418
BOS
1,097
VB
$193,932
BOS
$135,420
VB
2.04%
BOS
2.38%
| Metric | Virginia Beach | Boston | Difference |
|---|---|---|---|
| Total Spend | $193,932 | $135,420 | VB spent 43% more |
| Total Conversions | 418 | 1,097 | BOS 2.6x more |
| Account CPL | $464 | $123 | VB 3.8x higher |
| Monthly Spend | $13,852 | $9,673 | VB 43% higher |
| Monthly Leads | 30 | 78.4 | BOS 2.6x more |
| Total Clicks | 20,519 | 37,408 | BOS 82% more clicks |
| Conversion Actions | 41 (9 Primary) | 2 (2 Primary) | VB has 20.5x more |
| Managed By | Corporate | Outside Agency | — |
| Date Range | Jan 25 – Feb 26 | Jan 25 – Feb 26 | Same period |
Campaign Breakdown
VB runs the corporate template: AI Max Search gets 52% of budget at $438 CPL. Boston doesn't have AI Max at all — it gives PMax 54% and lets it deliver 82% of conversions at $81 CPL.
PMax CPL
PMax % Budget
PMax % Conversions
PMax Spend
The AI Max Problem: VB gives AI Max Search 52% of budget ($161K) at $438 CPL — 3.4x more expensive than PMax. Boston doesn't run AI Max at all. This single campaign decision costs VB an estimated $80K+ in wasted spend over 14 months.
14-Month Trends
Boston's CPL stays flat between $115–$127 for 13 straight months. VB swings wildly from $313 to $713. Consistency is the hallmark of clean signals — the algorithm knows what it's optimizing for.
| Month | VB Spend | VB Leads | VB CPL | BOS Spend | BOS Leads | BOS CPL | CPL Gap |
|---|---|---|---|---|---|---|---|
| Jan 25 | $10K | 18.3 | $530 | $9K | 72.0 | $124 | +$406 |
| Feb 25 | $10K | 22.0 | $442 | $9K | 68.0 | $127 | +$315 |
| Mar 25 | $13K | 26.9 | $500 | $9K | 75.0 | $121 | +$379 |
| Apr 25 | $14K | 32.7 | $415 | $9K | 80.0 | $117 | +$298 |
| May 25 | $14K | 33.2 | $412 | $10K | 82.0 | $117 | +$295 |
| Jun 25 | $13K | 21.3 | $633 | $9K | 76.0 | $122 | +$511 |
| Jul 25 | $13K | 40.7 | $313 | $10K | 85.0 | $116 | +$197 |
| Aug 25 | $14K | 35.5 | $385 | $10K | 88.0 | $115 | +$270 |
| Sep 25 | $14K | 38.8 | $352 | $10K | 84.0 | $119 | +$233 |
| Oct 25 | $13K | 22.4 | $595 | $10K | 86.0 | $119 | +$476 |
| Nov 25 | $17K | 43.9 | $379 | $10K | 90.0 | $116 | +$263 |
| Dec 25 | $17K | 30.7 | $559 | $10K | 87.0 | $119 | +$440 |
| Jan 26 | $16K | 28.4 | $578 | $10K | 85.0 | $119 | +$459 |
| Feb 26 | $17K | 23.5 | $713 | $10K | 39.0 | $244 | +$469 |
Conversion Signal Quality
Virginia Beach has 41 conversion actions with 9 marked Primary. Boston has 2 actions, both Primary. This is the single biggest structural difference between the two accounts and the primary driver of the CPL gap.
26 Zero-Conversion Actions
26 of 41 actions produced zero conversions over 14 months. Only 'Opportunity — New' is meaningful (63 conv).
9 Primary Actions (8 are noise)
The algorithm receives 9 signals but only 1 produces actual leads. The other 8 dilute optimization with junk data.
Multiple Phone Tracking Overlaps
Phone call tracking fires from multiple sources, inflating reported conversions and misleading Smart Bidding.
Business Profile Actions
Google Business Profile interactions counted as Primary conversions — map views and direction requests are not leads.
Bid Strategy: Max Conv Value
Optimizes for conversion value, not conversion volume. In lead gen, every lead has equal value — this strategy chases phantom value signals.
Signal Purity: ~11% — Only 1 of 9 Primary actions produces real leads. The algorithm is 89% confused.
"Schedule Me" Button Clicks
2,250 total over 14 months. A user actively requesting a consultation — the highest-intent action possible.
Calls from Ads
310 total over 14 months. A user calling directly from the ad — another high-intent action.
Signal Purity: 100% — Both Primary actions are real leads. The algorithm knows exactly what to optimize for.
89%
of VB's Primary signals are noise
0%
of Boston's Primary signals are noise
3.8x
CPL difference ($464 vs $123)
Brand & Branded Search
Branded search — people searching for 'Closet Factory' by name — is the highest-intent, lowest-cost traffic any business can get. How each account captures and converts that traffic reveals a fundamental strategic difference.
VB Brand Spend
$21,157
Via AI Max Search (Broad)
BOS Brand Spend
$17,130+
Dedicated campaign + PMax
VB Brand CPL
$167
126.5 conversions
BOS Brand CPL
$133
128+ conversions
Brand terms run through AI Max Search · Broad Match · Max Conv Value
The Problem
VB has no dedicated branded campaign. Brand terms like "closet factory" are lumped into the AI Max Search campaign alongside generic and competitor terms — all on Broad Match with Max Conversion Value bidding. This means Google decides how much to bid on VB's own brand name using the same polluted signals it uses for everything else.
Brand Ad Group Performance
Top Branded Keywords
Branded Search Term Waste
"Closet factory virginia beach reviews" — a branded review query — spent $332 with zero conversions. Because broad match is used, Google matches brand queries to non-brand ad groups too, diluting the signal.
Separate campaign · Target Impression Share · Controlled spend
The Strategy
Boston runs a dedicated "Search (Branded)" campaign with Target Impression Share bidding — the correct strategy for branded terms. The goal isn't to maximize conversions from brand searches (PMax does that automatically), it's to defend the brand name from competitors like Closets by Design who overlap 69% of the time.
Branded Campaign Performance
Branded Terms Across All Campaigns
PMax automatically picks up branded queries and converts them at the account's best CPL. The dedicated branded campaign acts as a defensive layer — ensuring Closet Factory's name always appears when someone searches for it, even if PMax doesn't bid on that particular auction.
Why Target Impression Share?
Closets by Design has 31.27% impression share in Boston and overlaps 69% of the time, outranking CF 83% of the time. The branded campaign ensures CF appears for its own name searches — a defensive necessity, not a conversion play.
| Factor | Virginia Beach | Boston | Impact |
|---|---|---|---|
| Dedicated Brand Campaign | No | Yes | BOS isolates brand spend |
| Brand Bid Strategy | Max Conv Value | Target Imp. Share | BOS defends brand visibility |
| Brand Match Type | Broad | Phrase / Exact | BOS controls brand queries |
| Brand Spend | $21,157 | $3,331 (campaign) | VB overspends 6.3x on brand |
| Brand Conversions | 126.5 | 128+ (all campaigns) | Similar output, different cost |
| Brand CPL | $167 | ~$133 (blended) | BOS 20% cheaper |
| Brand Waste | $332 on 0-conv terms | $960 (32% rate) | Both have waste |
| Competitor Defense | None — no imp. share bid | Target Imp. Share | BOS protects brand name |
| PMax Brand Capture | Uncontrolled | Complementary | BOS uses PMax + campaign |
Virginia Beach spends $21,157 on branded traffic through an AI Max Search campaign using Broad Match and Max Conversion Value bidding. Boston spends $3,331 on a dedicated branded campaign and lets PMax handle the rest. Both markets get roughly the same number of branded conversions (~127), but VB pays 6.3x more for the dedicated brand effort.
The difference: Boston's outside agency understands that branded search is a defensive play, not a conversion play. You bid on your own name to prevent competitors from stealing it — not to maximize conversions (PMax does that automatically). VB's corporate template treats brand terms the same as every other keyword: broad match, value-based bidding, no isolation. The result is overspending on the cheapest traffic source while giving Google's algorithm no clear signal about what branded traffic is worth.
Keyword Strategy
Virginia Beach uses 100% Broad Match across all 74 keywords — the corporate template default. Boston uses a hybrid approach: 57% Phrase, 38% Broad, 4% Exact across 694 keywords. The mixed strategy gives the outside agency more control over which queries trigger ads.
74 keywords · All Broad · Corporate template
Risk: 100% Broad gives Google maximum freedom to match queries. With dirty conversion signals, the algorithm uses that freedom to find cheap, low-intent traffic — kitchen remodelers, furniture shoppers, DIY enthusiasts.
694 keywords · Phrase + Broad + Exact · Outside agency
Advantage: Phrase match constrains Google to queries that contain the keyword phrase. Combined with 973 negatives (2.7x more than VB), the outside agency controls which searches trigger ads while still allowing PMax broad discovery.
| Factor | Virginia Beach | Boston | Impact |
|---|---|---|---|
| Match Type Mix | 100% Broad | 57% Phrase / 38% Broad / 4% Exact | BOS has more query control |
| Total Keywords | 74 | 694 | BOS covers 9.4x more terms |
| Negative Keywords | 358 | 973 | BOS blocks 2.7x more junk |
| Search Waste Rate | 56% | 54% | Similar — both need work |
| Wasted Spend | $63,083 | $35,185 | VB wastes $28K more |
| Top Wasted Category | Kitchen/Bath/Furniture | Closet organizer/systems | VB attracts wrong industry |
Search Term Waste
Both markets waste over 50% of search spend on terms that produce zero conversions. The difference: VB wastes $63K on 28,982 zero-conversion terms. Boston wastes $35K. VB's waste is nearly double because it spends more and has less query control.
VB Waste Rate
56%
$63,083 wasted
BOS Waste Rate
54%
$35,185 wasted
VB Zero-Conv Terms
28,982
of 29,088 total
BOS Negatives
973
vs VB's 358
Pattern: Kitchen remodels, bath renovations, furniture shopping — VB's broad match + dirty signals attract the wrong industry entirely.
Pattern: Boston's wasted terms are still closet-related — organizers, systems, competitors. Wrong intent, but right industry. The waste is less damaging.
Fraud & Bot Exposure
Dirty conversion signals don't just waste money — they actively attract fraud. When the algorithm optimizes for cheap, easy-to-trigger actions, bots and click farms become the cheapest source of 'conversions.' Here's how VB and Boston compare on fraud vulnerability.
Page views, phone tracking fires, business profile clicks, zero-conversion actions — all triggerable without human intent.
Only 1 of 9 Primary actions requires real human effort. The algorithm is 89% blind to lead quality.
Optimizes for phantom value signals, not real lead volume. Bots can generate high 'value' by triggering multiple low-effort actions.
100% Broad gives Google maximum query freedom. Dirty signals tell it to find cheap traffic. The intersection is bot traffic.
Fewer blockers means more junk queries get through, which means more opportunities for bots to trigger cheap conversions.
'Schedule Me' requires navigating a form flow. 'Calls from ads' requires a real phone call. Both require sustained human effort.
Both Primary actions are real leads. The algorithm only learns from genuine consultation requests.
Optimizes for cost-per-acquisition, not value. The algorithm is constrained to find conversions within a cost target, not maximize phantom value.
57% Phrase match constrains queries. Clean signals tell the algorithm to find real leads. The intersection is qualified homeowners.
2.7x more blockers than VB. More junk queries are filtered before they can trigger any action.
Virginia Beach
Critical — Nearly every structural factor amplifies fraud risk
Boston
Low — Clean signals and constrained matching minimize bot exposure
The Compounding Effect: VB's fraud vulnerability isn't just one thing — it's the combination of dirty signals + broad match + wrong bid strategy + fewer negatives. Each factor amplifies the others. Boston avoids this compounding by getting the fundamentals right on every dimension.
Prescriptive Strategy
The audit identifies the problems. This section prescribes the fixes — priority-ordered, with projected savings. VB needs structural overhaul. Boston needs refinement.
4 fixes · Estimated annual savings: $69,000 – $91,000
The Problem
Brand terms like 'closet factory' are lumped into AI Max Search alongside generic and competitor terms — all on Broad Match with Max Conversion Value bidding. VB spends $21,157 on branded traffic at $167 CPL.
How to Fix It
Projected Impact
Branded CPL drops from $167 to ~$30–50. Same conversions, ~$12,000–15,000 saved over 14 months.
The Problem
Without negative keywords, AI Max Search will continue bidding on branded queries even after the dedicated campaign exists. Both campaigns compete against each other, driving up costs.
How to Fix It
Projected Impact
Eliminates cannibalization. AI Max Search stops wasting budget on queries the branded campaign handles at 1/5th the cost.
The Problem
VB has 9 Primary actions but only 1 ('Opportunity — New') produces real leads. The other 8 — YouTube views, Get Directions, Business Profile clicks, duplicate phone tracking — feed junk signals to Smart Bidding. The algorithm is 89% confused.
How to Fix It
Projected Impact
This is the single highest-impact change. Clean signals let the algorithm optimize for actual leads instead of map clicks and video views. Conservative estimate: CPL drops from $464 to $250–300.
The Problem
If VB wants to conquest competitor terms ('closets by design,' 'california closets'), they should be isolated in their own campaign — not mixed into AI Max Search where they pollute the algorithm's learning.
How to Fix It
Projected Impact
Isolates expensive competitor traffic from the main algorithm. Prevents high-CPL competitor queries from inflating the blended CPL and confusing Smart Bidding.
4 refinements · Estimated additional savings: $4,000 – $8,000/year
The Problem
'Closet factory boston' appeared in the general Search campaign at $484 CPL. Branded queries are leaking past the dedicated branded campaign into the expensive general campaign.
How to Fix It
Projected Impact
Branded queries stop triggering expensive general campaign ads. Estimated savings: $2,000–4,000/year.
The Problem
Boston spends $10K (8% of budget) on competitor conquesting at $463 CPL — nearly 6x the PMax CPL of $81. Over 14 months, this produced only 22 leads.
How to Fix It
Projected Impact
If competitor leads underperform, reallocating $5K to PMax could generate ~62 additional leads at $81 CPL instead of 11 at $463.
The Problem
Top generic terms ('custom closets' at $714 CPL, 'closet organizer' at $492 CPL) are expensive. Boston uses 57% Phrase Match, which is strong, but Exact Match on proven converters could reduce waste.
How to Fix It
Projected Impact
Reduces wasted spend on close-but-not-quite search term variations. This is optimization, not a structural fix — Boston's foundation is already sound.
The Problem
Currently, Smart Bidding optimizes for form submissions — but not all form submissions become paying customers. Without offline conversion data, the algorithm can't distinguish a $15K project from a tire-kicker.
How to Fix It
Projected Impact
This is the next level beyond clean Primary actions. The algorithm learns to find people who not only submit forms but actually buy. Google's own documentation calls this the 'gold standard' for lead gen.
No additional budget required — these are structural fixes only
| Change | Current Cost | Projected Cost | Annual Savings |
|---|---|---|---|
| Dedicated branded campaign | $21,157 at $167 CPL | ~$5,000 at $40 CPL | $12,000 – $15,000 |
| Clean Primary actions (algorithm re-learns) | $464 blended CPL | $250–300 CPL (conservative) | $50,000 – $70,000 |
| Branded negatives in AI Max Search | Cannibalization waste | Eliminated | $3,000 – $5,000 |
| Total Estimated Annual Savings | $69K – $91K | ||
Conservative projection: $464 → $200–250 CPL — a 46–57% reduction from structural fixes alone
Even after implementing all fixes, VB's projected CPL of $200–250 would still be higher than Boston's $123. The remaining gap comes from two factors that require deeper strategic changes:
AI Max Search ($161K at $438 CPL)
Boston doesn't run AI Max Search at all. VB gives it 52% of budget. Eliminating or dramatically reducing this campaign and reallocating to PMax would close most of the remaining gap. This requires a strategic decision about the corporate template.
YouTube TV ($18K, 0 Conversions)
$18,089 spent with zero conversions over 14 months. Boston doesn't run YouTube TV. Reallocating this budget to PMax at $130 CPL would generate ~139 additional leads. This is the easiest budget decision on the table.
The Verdict
| Dimension | Virginia Beach | Boston | Winner |
|---|---|---|---|
| Cost Per Lead | $464 | $123 | BOS |
| Branded Strategy | No campaign · Broad · $167 CPL | Dedicated · Imp. Share · $133 CPL | BOS |
| Monthly Leads | 30 | 78 | BOS |
| Monthly Spend | $13,852 | $9,673 | BOS |
| PMax CPL | $130 | $81 | BOS |
| PMax Budget Share | 32% | 54% | BOS |
| Conversion Actions | 41 (9 Primary) | 2 (2 Primary) | BOS |
| Signal Purity | 11% | 100% | BOS |
| Match Type Control | 100% Broad | 57% Phrase / 38% Broad | BOS |
| Negative Keywords | 358 | 973 | BOS |
| Bid Strategy | Max Conv Value | Target CPA | BOS |
| Fraud Vulnerability | 9.2 / 10 | 2.1 / 10 | BOS |
| CPL Stability | $313–$713 range | $115–$127 range | BOS |
| Leads per $1,000 | 2.2 | 8.1 | BOS |
| AI Max Search | $161K at $438 CPL | Does not exist | BOS |
| YouTube TV | $18K, 0 conversions | Does not exist | BOS |
16 – 0
Boston wins every measurable dimension.
This is not a close contest. Virginia Beach spends 43% more money to get 62% fewer leads at 3.8x the cost. The difference is entirely structural: clean conversion signals, mixed match types, PMax-first budget allocation, and a bid strategy designed for lead generation. The outside agency running Boston got every fundamental right that the corporate template gets wrong.