META LEAD GENERATION

Home-improvement lead generation: balancing cost, volume and lead quality

The lowest-cost creative and the highest-volume creative were not the same—so the scale decision had to include lead quality.

UAE · July 2026Anonymised case studyReported results with context

THE DECISION IN VIEW

The lowest-cost creative and the highest-volume creative were not the same—so the scale decision had to include lead quality.

Reported snapshot

33

Meta leads

AED 38.48

Campaign cost per lead

AED 27.36

Best static cost per lead

22

Leads from the reel

Context

A premium flooring campaign needed service-specific enquiries without fragmenting a limited budget across too many audiences. The offer centred on helping property owners move from interest to a relevant assessment.

Diagnosis

The campaign created two different signals. A static produced the lowest cost per lead, while a reel produced most of the volume at a higher AED 40.91 CPL. Declaring either format the winner would have ignored the trade-off between efficiency and demand.

Decisions

  1. Keep the service promise specific

    Campaign language stayed close to flooring categories and the assessment people were being invited to request.

  2. Give each creative a measurable job

    The static remained the efficiency control while the reel was evaluated as the volume route.

  3. Avoid unnecessary audience splits

    The available budget was concentrated so format and message differences could be read more clearly.

  4. Move the decision downstream

    Qualified enquiries and booked visits were added to the reporting priorities before reallocating budget.

Result and learning

The mid-month snapshot recorded 33 Meta leads at AED 38.48 per lead. The best static produced leads at AED 27.36, while the reel generated 22 leads at AED 40.91.

A later reporting period recorded 49 form leads and 40 message conversations. Those totals are kept separate because they represent different periods and conversion actions. The durable lesson was to protect the efficient control while comparing volume against contactability, service fit and booked assessments.

Figures are platform-reported. The primary snapshot began mid-month; the later 49 form leads and 40 messages are separate result types from a later period. Leads and messages are not confirmed site visits or won projects. Client identity is intentionally withheld. Past performance does not guarantee future results.