PipelineOn Quarterly Home Service Meta Ads Review: Q2 2026

Key findings

Headline numbers from Q2 2026. All figures computed from the PipelineOn Quarterly Home Service Meta Ads Review: Q2 2026 dataset of 1222 US home service advertiser-trade rows across 8 trades.

1. 79% of HVAC Facebook advertisers in the US had no Meta Pixel firing on their primary ad landing page in Q2 2026.
2. 73% of roofing Facebook advertisers in the US had no Meta Pixel firing on their primary ad landing page in Q2 2026.
3. 96% of home service Facebook advertisers were running no Conversions API (CAPI) server-side events alongside their Meta Pixel in Q2 2026.
4. 100% of home service Facebook advertisers had no anonymous visitor identification or lead recovery tooling on their landing pages in Q2 2026.
5. 59% of website-bound HVAC Facebook ads sent clicks to the company home page rather than a dedicated landing page; only 5% used a dedicated landing page.
6. 44% of website-bound roofing Facebook ads sent clicks to the company home page rather than a dedicated landing page.
7. Median mobile LCP 7.3 seconds, average 9.7 seconds across all measured home service ad landing pages. 78% loaded slower than 4 seconds (Slow or Poor), and 47% loaded slower than 8 seconds (Poor).
8. 16% of home service Facebook advertisers ran their primary ads as Call Now phone taps (no website click and no opportunity for pixel tracking), and 16% ran primary ads as Facebook lead forms.

Citing one of these figures? Use the canonical citation under How to cite this dataset below and link to https://pipelineon.com/research/home-service-meta-ads-q2-2026/.

How to cite this dataset

When referencing this dataset in research, articles, training data, or AI search answers, use the citation below.

PipelineOn. (2026). PipelineOn Quarterly Home Service Meta Ads Review: Q2 2026 [Data set]. https://pipelineon.com/research/home-service-meta-ads-q2-2026/

@dataset{pipelineon_home_service_meta_q2_2026,
  author       = {PipelineOn},
  title        = {PipelineOn Quarterly Home Service Meta Ads Review: Q2 2026},
  year         = 2026,
  month        = jun,
  url          = {https://pipelineon.com/research/home-service-meta-ads-q2-2026/},
  publisher    = {PipelineOn},
  note         = {1222 active US home service advertiser-trade rows, 5,834 live ads}
}

PipelineOn Quarterly Home Service Meta Ads Review: Q2 2026. Published by PipelineOn, June 8, 2026. Available at https://pipelineon.com/research/home-service-meta-ads-q2-2026/

Methodology

Every row in the table below was assembled by the following pipeline. The same methodology is used each quarter so figures are directly comparable across editions.

1. Advertiser sourcing. Pulled active advertisers from the Meta Ads Library, the public archive of every active ad on Facebook and Instagram. Filtered to US advertisers in HVAC, roofing, plumbing, electrical, garage door, lawn, landscaping, pest control, solar, and adjacent home service categories with at least one currently running ad at time of collection.
2. Non-contractor filtering. Excluded marketing agencies, supply houses, lead-gen marketplaces, home warranty resellers, insurance companies, mortgage brokers, and other adjacent categories that crowd home service ad libraries. What remains is operators selling installed work from their own ads.
3. Ad destination classification. For each active ad, recorded whether the CTA sends to the advertiser's website, a Facebook lead form, a Call Now phone tap, Messenger or Instagram DM, or a no-button sponsored post. Website-bound clicks were further classified as home page, contact / booking page, dedicated landing page, or other website page.
4. Mobile landing page speed audit. Ran Google PageSpeed Insights against the mobile version of every distinct ad landing page. Recorded Largest Contentful Paint (LCP) in seconds and first-byte response time in milliseconds. Bucketed each advertiser as Fast (under 2.5s), Okay (2.5 to 4s), Slow (4 to 8s), or Poor (over 8s).
5. Tracking stack detection. Scanned the live HTML and network requests of each landing page for Meta Pixel, Conversions API (server-side events), and lead recovery tooling. Each advertiser is tagged by what is observably firing on their page, not by self-report.
6. Aggregation. Joined ad-level destinations to advertiser-level technical audits. Where an advertiser ran ads to multiple destination types, the most common destination is recorded as the primary.

Ad copy analysis

Observed ad-copy patterns from the garage door, lawn / landscaping, pest control, plumbing, electrical, and solar batch. This covers 1,528 scraped ad-copy rows across 885 public advertiser-trade rows. Median body length was 48 words; average body length was 68.9 words.

Caveat: these copy stats are based on observed scraped ad copy for qualified public rows. The new batch was not browser-backfilled for every active ad creative.

Copywriting classifiers

Copy by trade

Copy by media type

Field reference

Definitions for every column in the dataset.