Melleka Marketing

Agency Handoff
Performance Analysis

A detailed investigation into why ROAS dropped during the April 14–15 transition, who made what changes, and what happened.

Client
Vegamour
Platform
Google Ads
Analysis Period
April 14 vs April 15, 2026
Prepared By
Melleka Marketing
01

The Headline Numbers

Here is the day-over-day comparison between April 14 (the previous agency's last day) and April 15 (Melleka's first day managing the account).

Total Spend
$8,327
+$2,460 (+42%) vs yesterday's $5,867
Primary ROAS
0.90x
Down from 1.93x (-53%)
All-Conv ROAS
5.09x
Down from 11.75x (-57%)
Primary Conv Value
$7,503
Down from $11,315 (-34%)
All-Conv Value
$42,419
Down from $68,920 (-38%)
Clicks
2,101
Down from 2,154 (-2%)
The Key Question
Spend went UP 42% while revenue went DOWN 34%. Why? The answer lies in a single campaign — and a single change made by the previous agency on their way out.
02

Root Cause: Branded Search

One campaign absorbed an extra $3,500 in spend with diminishing returns. This single campaign accounts for nearly the entire ROAS drop.

Apr 14 — Previous Agency

Branded Search Spend
$1,315
All-Conv Revenue
$36,217
All-Conv ROAS
27.54x

Apr 15 — After Their Changes

Branded Search Spend
$4,848
+269% ($3,533 more)
All-Conv Revenue
$27,661
-$8,556 less revenue
All-Conv ROAS
5.70x
Down from 27.54x (-79%)
What This Means
The previous agency's Branded Search campaign was running at 27.54x ROAS on $1,315/day — an excellent result. After they changed the bidding strategy (detailed in the next section), Google spent 3.7x more on the same campaign with significantly LESS revenue. The extra $3,533 in spend was essentially wasted on diminishing-return branded clicks.
03

Branded Search: 8-Day Trend

The daily spend and ROAS for Branded Search USA over the past 8 days. Notice the spike on April 15 — the day after the bidding strategy was changed.

Apr 8
$4,141
14.3x
Apr 9
$2,448
17.2x
Apr 10
$2,417
16.9x
Apr 11
$2,858
15.6x
Apr 12
$1,934
20.5x
Apr 13
$1,557
28.0x
Apr 14
$1,315
27.5x
Apr 15
$4,848
5.7x
The Pattern Is Clear
When Branded Search spent $1,300–$1,900/day, ROAS was 20x–28x. When it spent $4,800+, ROAS crashed to 5.7x. There's a natural ceiling on how many brand searches happen each day — past that ceiling, Google starts bidding on lower-quality, higher-cost clicks. This campaign performs best when lean.
04

Exactly What Changed & Who Did It

A complete, timestamped audit trail from the Google Ads change history — every change, every user, every device. All times in UTC.

Previous Agency Changes (neighborhoodreachmarketing@gmail.com)

Apr 14, 8:24 AM UTC
Previous Agency — Web Client
Raised NB-Broad+AI-Max budget from $600/day to $850/day — a 42% increase on a campaign that was already underperforming (0.22x ROAS over 30 days).
Apr 14, 11:37 AM UTC
Previous Agency — Web Client
Tripled Remarketing DG budget from $200/day to $600/day — this campaign had a 0.61x ROAS (losing money on every dollar spent).
Apr 14, 11:40 AM UTC
Previous Agency — Web Client
Enabled 5 previously paused campaigns: NB-DSA-Hair, Nonbranded Exact USA, NB-Hair Growth Serum, NB-Gummies, NB-Problems. These campaigns were paused for a reason — they were all underperformers.
Apr 14, 6:05 PM UTC
Previous Agency — Mobile App
THE CRITICAL CHANGE: Switched Branded Search bidding from Max Conversions (with $24 CPA cap) to Max Conversion Value (with 4.4x tROAS). This fundamentally changed how Google spends money on this campaign. A CPA cap restricts spend; a tROAS target lets Google spend as much as it wants as long as it believes it can hit the return target. This single change caused the spend explosion.
Apr 14, 6:43 PM UTC
Previous Agency — Recommendations
Auto-applied Google's keyword recommendations — added new keywords like "best castor oil for eyelashes", "hairtamin gummies", "genetic hair loss treatment" to NB-Exact campaign. These irrelevant keywords dilute targeting quality.
Apr 14, 10:10 PM UTC
Previous Agency — Web Client
Additional campaign-level changes on NB-DSA-Hair (exact nature unclear from log).
Apr 15, 12:03 AM UTC
Previous Agency — Mobile App
Lowered Branded Search tROAS from 4.4x to 3.5x — making the target EVEN LOOSER. A lower tROAS target tells Google "I'm OK with lower returns" which allows even more aggressive spending.
Apr 15, 12:08–12:13 AM UTC
Previous Agency — Web Client
Changed bidding strategies on 6 more campaigns: NB-DSA-Hair (set tROAS to 1.0x), NB-Brow (moved to portfolio strategy), NB-Gummies (moved to portfolio strategy), NB-Hair Growth Serum (moved to portfolio strategy), Shopping NB Hair Growth Supplements (switched to Enhanced Manual CPC), Nonbranded Shopping (moved to portfolio strategy). A flurry of strategy changes at midnight before handoff.

Melleka Marketing Changes

Apr 15, 4:34–4:40 PM UTC
Bryan @ Melleka — Web Client
Cut budgets on underperforming campaigns: NB-Brow ($400→$200→$100), NB-HYDR8 ($380→$100), NB-Hair-Shampoo ($200→$100), NB-Problems ($130→$50). Also paused 5 money-losing campaigns: NB-DSA-Hair, NB-DSA, NB-Broad+AI-Max, Remarketing DG, NB-Hair Growth Serum. Bumped Branded Search budget from $4,500 to $5,000 (minor — the bidding strategy change was already causing the overspend).
Apr 15, 4:39 PM UTC
David @ Melleka — Web Client
Created 4 new strategically structured campaigns: "Search | Branded", "Search | Non-Branded", "Search | Competitor Conquesting", "Search | Problem-Aware" — with properly segmented ad groups, fresh keywords, and clean campaign architecture.
Apr 15, 7:39 PM UTC
Anthony @ Melleka — Web Client
Increased NB-Brow budget from $100 to $1,000 for testing purposes.
05

Full Campaign Comparison

Every campaign that spent money on either day, with side-by-side performance.

Campaign Apr 14 Spend Apr 15 Spend Apr 14 All-ROAS Apr 15 All-ROAS Status Today
Branded Search USA $1,315 $4,848 27.54x 5.70x Enabled
PMax Hair Growth Serums $1,317 $999 10.69x 6.27x Enabled
Nonbranded Shopping USA $1,252 $352 1.97x 3.04x Enabled
Branded Shopping USA $1,074 $849 9.66x 5.70x Enabled
NB-Lash Search $40 $517 3.87x 3.15x Enabled
NB-Brow Search $465 $62 1.76x 0.00x Enabled
PMax Shampoo Kits $357 $109 7.81x 7.63x Enabled
NB-Exact Search $222 $325 3.66x 0.00x Enabled
NB-DSA-Hair (paused by M.M.) $268 $0 1.38x — Paused
NB-DSA (paused by M.M.) $170 $387 2.10x 0.86x Paused
Remarketing DG (paused by M.M.) $146 $0 2.54x — Paused
NB-Broad+AI-Max (paused by M.M.) $86 $0 0.10x — Paused
NB-Hair-Shampoo (paused by M.M.) $75 $30 1.04x 5.50x Paused
TOTAL $5,867 $8,327 11.75x 5.09x
Campaigns Melleka Paused (Green Rows)
Every campaign Melleka paused was either losing money (ROAS below 1.0x) or barely breaking even. Notably, NB-Broad+AI-Max had a 0.10x ROAS — spending $10 for every $1 returned. The previous agency had actually INCREASED its budget to $850/day the morning of handoff.
06

Why the Bidding Strategy Change Matters

This is the technical explanation of what happened to Branded Search and why it caused the ROAS crash.

Before (How It Was Running)
Strategy: Maximize Conversions with $24 Target CPA

How it works: Google tries to get as many conversions as possible while keeping the average cost per conversion around $24. This acts as a spending brake — if clicks get too expensive, Google stops bidding.

Result: Controlled, efficient spend. $1,315/day with 27.54x ROAS. The campaign only captured high-quality, high-intent brand searches.
After (What They Changed It To)
Strategy: Maximize Conversion Value with 3.5x Target ROAS

How it works: Google tries to generate as much revenue as possible while maintaining a 3.5x return. There is no spending brake — Google can spend as much as it wants as long as it believes it can hit 3.5x.

Result: Unconstrained spending. $4,848/day with only 5.70x ROAS. Google aggressively bid on lower-quality searches to try to maximize total revenue, but with rapidly diminishing returns.
The Timing Is Important
The previous agency made this change at 6:05 PM on April 14 from their mobile phone, then lowered the target ROAS even further to 3.5x at 12:03 AM on April 15 (also from their phone). Then at 12:08–12:13 AM, they changed bidding strategies on 6 additional campaigns from their web client. This was a rapid sequence of changes made in the final hours before handoff — without documentation or explanation.
07

What Melleka Did on Day 1

Despite inheriting an account with freshly changed bidding strategies and inflated budgets, the Melleka team took immediate corrective action.

Campaigns Paused
5
Stopped money-losing campaigns
Budgets Cut
4
Reduced waste on underperformers
New Campaigns Built
4
Clean, strategic architecture

Campaigns We Paused (All Were Losing Money)

Campaign 30-Day ROAS 30-Day Spend Why We Paused
NB-Broad+AI-Max 0.22x $5,300 Spending $10 for every $1 returned
Remarketing DG 0.61x $13,600 Losing money on remarketing
NB-DSA 0.73x $2,800 Dynamic search ads underperforming
NB-DSA-Hair 1.04x $3,200 Barely breaking even
NB-Hair Growth Serum 0.89x $1,100 Below breakeven

New Campaigns We Built

New Campaign Strategy Purpose
Search | Branded || Structured branded capture Take over brand search with clean segmentation
Search | Non-Branded || High-intent generic terms Capture "hair growth serum" searchers efficiently
Search | Competitor Conquesting || Competitor brand targeting Win customers searching for competitors
Search | Problem-Aware || Symptom-based targeting Reach people searching for hair loss solutions
08

The Verdict

A clear summary of what caused the ROAS drop and who is responsible.

What the Previous Agency Did Wrong
1
Changed the Branded Search bidding strategy to remove the spending brake. Switching from Max Conversions ($24 CPA cap) to Max Conversion Value (3.5x tROAS) gave Google unlimited permission to spend. This caused Branded Search to go from $1,315/day to $4,848/day with diminishing returns.
2
Made these changes at the last possible moment. The critical bidding change was made at 6:05 PM on their final day (from a phone), then adjusted again at midnight. Six additional campaigns had their strategies changed between 12:08–12:13 AM.
3
Enabled 5 paused campaigns and tripled the Remarketing budget on their last day. NB-DSA-Hair, Nonbranded Exact, NB-Hair Growth Serum, NB-Gummies, and NB-Problems were all re-enabled. Remarketing DG went from $200/day to $600/day despite losing money at 0.61x ROAS.
4
Auto-applied Google's keyword recommendations without review. Added irrelevant keywords like "hairtamin gummies" and "best castor oil for eyelashes" that dilute campaign targeting quality.
What Melleka Did Right
1
Immediately paused 5 money-losing campaigns that were dragging down overall ROAS. Combined 30-day ROAS of paused campaigns: 0.55x (losing money).
2
Cut budgets on 4 additional underperformers to stop the bleeding while evaluating the account.
3
Built 4 new, strategically segmented campaigns with clean architecture: Branded, Non-Branded, Competitor Conquesting, and Problem-Aware — each with targeted ad groups and fresh keyword research.
4
Identified the root cause within hours. The bidding strategy change on Branded Search was detected and documented. A plan to fix it is ready for implementation.
09

Recommended Next Steps

Immediate actions to restore ROAS to pre-handoff levels and set the account up for growth.

Priority 1 — Fix Branded Search (Immediate)
Revert Branded Search bidding strategy back to Maximize Conversions with a CPA cap, or set a much tighter tROAS target (10x+). This single change will bring Branded Search back to its $1,300–$2,000/day sweet spot and restore 20x+ ROAS.
Priority 2 — Ramp New Melleka Campaigns (This Week)
The 4 new campaigns (Branded, Non-Branded, Competitor Conquesting, Problem-Aware) are built and ready. Once Branded Search is fixed, we'll activate these with controlled budgets and proper conversion tracking to systematically find profitable non-branded growth.
Priority 3 — Full Account Audit (Next 7 Days)
Deep-dive every campaign, ad group, and keyword. Fix conversion tracking (currently 9+ primary conversion actions causing double counting). Establish proper ROAS benchmarks and budget allocation based on actual profitability data.
Prepared by Melleka Marketing — Confidential — April 2026