Skip to content
Kuraib Ali
Selected work

Dubai aesthetics clinic (confidential)

Fourteen months with an agency. 847 broad-match keywords. Not one negative.

Sector
Healthcare / aesthetics · Dubai
Timeframe
28 days
Engagement
Google Ads rebuild: tracking, negative keywords, bidding recalibration

The situation

A Dubai aesthetics clinic had run Google Ads with a previous agency for fourteen months on an AED 18,000 monthly budget.

Bookings were coming in, but cost per lead had been rising and nothing could be verified, because enhanced conversions were never configured and all tracking was browser-side. The account audit found 847 keywords live in broad match with no negative keyword list, Performance Max mixing brand and treatment queries in one campaign, and smart bidding aimed at a Target CPA of AED 90 — a placeholder from the previous agency — when the verified cost per lead over six months was AED 134. The algorithm had been under-bidding every auction.

The move

  1. 01

    I verified the signal first: a GTM server container, enhanced conversions configured with hashed email from the booking form, checked against Google Ads conversion diagnostics before any campaign change.

  2. 02

    I built the negative keyword architecture from fourteen months of search-term data — 312 negatives across six lists covering informational queries, competitor brands, treatments the clinic does not offer and areas outside its radius — and moved broad-match keywords to phrase and exact where intent was clear.

  3. 03

    I recalibrated Target CPA to AED 130, close to the verified history, so the algorithm could win real auctions; separated Performance Max asset groups by treatment; and split brand from non-brand with budget controls.

What changed

States of the system, not metrics — these are things that either were or were not true, so they can be stated without claiming a result.

Before and after states for the Dubai aesthetics clinic (confidential) engagement.
BeforeAfter
Conversion trackingBrowser-side only, unverifiableServer-side enhanced conversions, verified
Keywords847 broad match, zero negativesPhrase and exact where clear, 312 negatives in six lists
Smart bidding targetAED 90 placeholder vs AED 134 realAED 130, calibrated to verified data
Performance MaxBrand and treatments mixedAsset groups by treatment, brand separated

Scroll sideways to compare →

What shipped

  • GTM server container + enhanced conversions, verified in diagnostics
  • 312 negative keywords across six lists, from 14 months of search terms
  • Broad → phrase and exact match where intent was clear
  • Target CPA recalibrated to verified history
  • Performance Max asset groups by treatment; brand and non-brand separated

The scorecard

What this engagement was held to, and what the platform reported. Each figure names the source it was pulled from.

Cost per lead
AED 115 target reached
Spend ÷ platform-verified booking-form conversions, Google Ads
Search impression share
31% → 58%
Google Ads auction insights
Wasted spend
−44%
Spend on non-converting queries, first month after negatives
Time to target
28 days
Days from structural rebuild to verified CPL

✓ Every figure is sourced from platform reporting — Meta Ads Manager, Meta Events Manager, Google Ads or the client’s CRM — never modelled or estimated. The full write-up, as published, follows below.

Full write-upGoogle Ads for Dubai Clinics: Strategy, Results & Verified CPL | Case Study
Quick Answer

Rising Google Ads CPL after a previous agency engagement is frequently a keyword-hygiene and tracking problem, not a bidding problem, this case shows a Dubai clinic’s account with 847 broad-match keywords and no negative keyword list reaching a verified AED 115 CPL target within 28 days of a structural rebuild.

01. The Situation

Previous agency. High spend. No server-side tracking. No negative keywords.

A Dubai aesthetics clinic had been running Google Ads with a previous agency for 14 months. The account had a AED 18,000 monthly budget and was generating bookings, but CPL had been rising steadily and the clinic could not verify attribution because enhanced conversions were not configured. All conversion tracking was browser-side only.

Account audit revealed 847 keywords active in broad match with no negative keyword list. Performance Max campaigns running without asset group segmentation, mixing treatment-specific and brand queries in the same campaign. Smart bidding set to Target CPA using a placeholder number from the previous agency’s estimate, not from actual verified conversion data.

02. The Audit Findings

Four structural problems. All fixable in the first two weeks.

  1. 01

    847 active keywords in broad match with zero negative keyword list, budget being consumed by irrelevant queries including competitor brand searches and informational queries with no commercial intent

  2. 02

    Enhanced conversions not configured, all tracking browser-side, estimated 30–40% signal loss from iOS and browser restrictions

  3. 03

    Smart bidding Target CPA set to AED 90 based on a previous agency estimate, actual verified CPL from the past 6 months of data was AED 134, meaning the algorithm was systematically under-bidding and losing auctions

  4. 04

    Performance Max campaign mixing brand and non-brand traffic with no asset group separation, making it impossible to evaluate treatment-specific performance or control spend allocation

03. What Was Done

Tracking verified. Negatives built. Bidding recalibrated.

GTM Server Container deployed. Enhanced conversions configured with hashed email data from the booking form. Verified against Google Ads conversion diagnostics before any campaign changes were made. Signal quality confirmed before touching bid strategy.

Negative keyword list built from 14 months of search term data: 312 negative keywords added across 6 negative lists covering informational queries, competitor brands, treatment types not offered by the clinic, and geographic areas outside the clinic’s service radius. Broad match keywords migrated to phrase and exact match where query intent was clear.

Smart bidding Target CPA recalibrated to AED 130, close to the verified historical CPL, allowing the algorithm to bid competitively in real auctions. Performance Max asset groups separated by treatment category. Brand and non-brand campaigns separated with budget controls.

04. The Result

Verified CPL target achieved within 28 days of structural rebuild.

CPL from Google Ads reached the AED 115 verified target within 28 days of the rebuild. Search impression share increased from 31% to 58% as recalibrated bidding began winning competitive auctions. Wasted spend on irrelevant queries dropped by 44% in the first month as negative keywords eliminated non-converting traffic.

AED 115Verified CPL achieved
58%Search impression share
−44%Wasted spend reduction
28 daysTime to CPL target
✓ All figures sourced from Google Ads platform reporting. CPL calculated as total spend divided by platform-verified booking form conversion count for the stated periods. Enhanced conversions verified via Google Ads conversion diagnostics before reporting.
← Real Estate CPL Dubai← E-Commerce ROAS← Healthcare Bookings← KSA Investor LeadsAll Case Studies →
What This Demonstrates

Google Ads performance depends on verified conversion data before smart bidding is trusted

Smart bidding is powerful when given accurate conversion data. When the Target CPA is set to a placeholder figure rather than a verified historical CPL, the algorithm systematically mis-bids every auction. This account had 14 months of campaign history that was instructive once the tracking was verified and interpreted correctly. The negative keyword architecture built from that history prevented thousands of AED in wasted spend from the first week of the restructure.

See Meta & Google Ads Service →
Frequently Asked

What was wrong with the account before this engagement?

The account had 847 active keywords in broad match with zero negative keywords, consuming budget on irrelevant searches, alongside no server-side conversion tracking, meaning bidding decisions were being made on incomplete data after 14 months with a previous agency.

What was fixed first?

Tracking was corrected before any bidding changes were made. A GTM Server Container was deployed and enhanced conversions were configured with hashed email data from the booking form, verified against Google Ads’ own conversion diagnostics before campaign structure was touched.

How quickly did CPL improve after the rebuild?

CPL reached the AED 115 verified target within 28 days of the structural rebuild. Search impression share increased from 31% to 58% as recalibrated bidding began winning more competitive auctions with cleaner budget allocation.

By Kuraib Ali, Search & Growth Consultant, Dubai · Meta Blueprint · Google Ads Certified  ·  Last updated: August 3, 2026

Next project

Lahore property agency (confidential)

Hundreds of clicks a week in Ads Manager. Silence in WhatsApp.

Working on something with a similar shape?

admin@kuraib.site