# Meta Ads Benchmarks for B2B SaaS > No 2026 Meta ads benchmark has a software row. The 36 category rows, the nearest fits for a B2B SaaS, and why the published medians do not multiply together. Source: https://saasadsstudio.com/research/meta-ads-benchmarks-for-b2b-saas/ Author: The SaaS Ads Studio Published: 2026-10-07 Updated: 2026-10-07 --- ## Key figures - **0**: Software or SaaS rows among the 36 category rows in the 2026 Facebook ads benchmark set - **$27.39**: Median cost per lead, US Facebook leads campaigns, all industries, USD - **$35.87**: Median cost per lead for Industrial & Commercial, the nearest row to a B2B SaaS, USD - **$21.08**: Cost per lead derived on this page from median cost per click and median conversion rate, USD - **+12%**: Meta average price per ad, Q2 2026, year on year (Meta's own metric, all advertisers) If you sell software and run Meta ads, the benchmark you will be pointed to has 36 category rows and none of them is yours. This page records what the 2026 set measures, which rows sit nearest a B2B SaaS, and which of its numbers do not hold together, so it can be cited with its limits attached. ## The 2026 headline figures {#headline} The dataset is WordStream by LocaliQ's Facebook Ads Benchmarks 2026, last updated 14 September 2026. Read on 7 October 2026, campaigns using the traffic objective show a median click-through rate of 1.93% and a median cost per click of $0.60. Campaigns using the leads objective show a median click-through rate of 2.70%, a cost per click of $1.80, a conversion rate of 8.54% and a cost per lead of $27.39. Every money figure is in US dollars, and every figure the report calls an average is a median. The same numbers are carried by [PPC Land's write-up of the 2026 edition](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/), published 14 September 2026 and read on 7 October 2026. Against the 2025 edition, the publisher reports traffic cost per click down 14.29% from $0.70 and cost per lead down 0.98% from $27.66, both in US dollars. [Digital Applied carries the 2025 value beside every 2026 headline figure](https://www.digitalapplied.com/blog/facebook-ads-benchmarks-2026-cpc-cpm-ctr-industry), updated 22 September 2026 and read on 7 October 2026. WordStream and LocaliQ are one company publishing one dataset, so they are never two sources for one number. WordStream is also on this site's competitor list, so we name it with the date and do not link it: the page is at wordstream.com/blog/facebook-ads-benchmarks-2026. The campaigns are drawn from the publisher's own customers, which makes this a vendor describing its own client base rather than a neutral survey of Meta advertisers. ## 36 category rows, none of them software {#categories} ![All 36 category rows in the 2026 Facebook ads benchmark set, traffic cost per click and leads cost per lead in US dollars, with the missing software and SaaS row drawn as an empty slot](/images/blog/meta-ads-benchmarks-for-b2b-saas-categories.svg "Every row in both tables, so the absence of a software row can be checked rather than taken on trust") The traffic table has 22 categories and the leads table has 14, so the set has 36 category rows (22 plus 14). Not one is software, SaaS or technology. Every row and its median is reproduced in [Digital Applied's industry tables](https://www.digitalapplied.com/blog/facebook-ads-benchmarks-2026-cpc-cpm-ctr-industry), read on 7 October 2026. What fills the list is local and consumer business: dentists, physicians and surgeons, real estate, restaurants and food, furniture, two automotive categories. That is the same shape as the publisher's search benchmarks, which have 23 categories and no software row either, as [we recorded on 1 October 2026](https://saasadsstudio.com/research/why-google-ads-benchmarks-miss-saas/). A B2B SaaS buying Meta ads in 2026 has no published row on either platform from the dataset most pages quote. ## The nearest rows to a B2B SaaS {#nearest} Business Services appears only in the traffic table, with a median click-through rate of 1.64% and a median cost per click of $0.70 in US dollars, as listed by [Digital Applied](https://www.digitalapplied.com/blog/facebook-ads-benchmarks-2026-cpc-cpm-ctr-industry) on 7 October 2026. It has no leads row, so there is no published cost per lead for it at all. Industrial & Commercial is the only category in both tables with a business buyer in it. On the traffic objective its median click-through rate is 2.06% and its cost per click $0.54. On the leads objective its click-through rate is 2.57%, cost per click $1.55, conversion rate 4.50% and cost per lead $35.87. All money is in US dollars, and every figure is from [Digital Applied's tables of the 2026 edition](https://www.digitalapplied.com/blog/facebook-ads-benchmarks-2026-cpc-cpm-ctr-industry), read on 7 October 2026. That row is also volatile. The publisher reports its leads conversion rate down 51.77% year on year and its traffic cost per click down 37.21%, both carried by [PPC Land](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/) and read on 7 October 2026. $35.87 is 31% above the all-industry median: derived on this page, 35.87 divided by 27.39 is 1.3096, both in US dollars and both from the 2026 edition [as carried by Digital Applied](https://www.digitalapplied.com/blog/facebook-ads-benchmarks-2026-cpc-cpm-ctr-industry), read on 7 October 2026. Even then, a lead on this objective is a form submission, often on a form inside Meta's own apps, while a SaaS account usually counts a trial or a demo request on its own site. ## How thin the sample is {#sample} The traffic data is 1,377 US campaigns run between 1 April 2025 and 30 June 2026, with at least 11 campaigns per category. The leads data is 452 US campaigns with at least four per category. Both were read off the publisher's about-the-data note on 7 October 2026 and are carried by [PPC Land](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/). Added together, 1,377 and 452 make 1,829 campaigns, which the report's opening describes as almost 1,800. Two oddities in that note are worth knowing before citing it, and [PPC Land noted both](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/). The leads window is printed as ending 30 June 2028, a date that has not happened and presumably means 2026. And both samples are described as search advertising campaigns, wording that appears to be carried over from the search edition. A third is worth adding: the editions overlap. The 2025 edition covered April 2024 to June 2025, as [PPC Land reported on 9 September 2025](https://ppc.land/facebook-ad-costs-jump-21-as-lead-campaigns-struggle-while-traffic-ads-thrive/), and the 2026 traffic window starts on 1 April 2025. Both editions therefore contain April, May and June 2025, three months of each 15-month window, which is 20% (3 divided by 15, derived on this page). A year-on-year change between them compares two windows that share a fifth of their calendar. The 2025 edition's own note, read on 7 October 2026 at wordstream.com/blog/facebook-ads-benchmarks-2025, gives 554 traffic and 726 leads campaigns, so the leads sample shrank between editions while the traffic sample grew. ## Two numbers in the report that do not hold together {#derived} ![A 1.80 US dollar median cost per click divided by an 8.54 per cent median conversion rate gives 21.08 US dollars per lead against a published 27.39, and the conversion rate change from 7.72 to 8.54 per cent works out at a rise of 10.6 per cent](/images/blog/meta-ads-benchmarks-for-b2b-saas-medians.svg "Both derived figures beside the published ones, so each step of the arithmetic can be checked") $21.08 per lead, in US dollars: derived on this page, not measured. The inputs are the leads median cost per click of $1.80 and the median conversion rate of 8.54%, both from the 2026 edition, read on 7 October 2026 and carried by [PPC Land](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/). The arithmetic is 1.80 divided by 0.0854, which is 21.0773, so $21.08. The published median cost per lead is $27.39, about 30% higher (27.39 divided by 21.0773 is 1.2995). That gap is not a mistake. Each median is taken separately across campaigns, so the median click price and the median conversion rate do not come from the same campaign and do not multiply back into the median cost per lead. PPC Land ran the same check. On the publisher's search edition the same arithmetic landed within 0.6% of the published figure, as [our Google page shows](https://saasadsstudio.com/research/why-google-ads-benchmarks-miss-saas/). On this Meta set it does not, so a forecast built from cost per click and conversion rate would understate cost per lead here. Plan from the cost per lead row. The conversion rate change is stated two ways. The report's summary says conversion rate rose by almost 11%; its conversion rate section says it fell 0.98%, which is the same figure it gives for cost per lead. The 2025 edition published 7.72%, read on 7 October 2026 and carried by [PPC Land's 2025 write-up](https://ppc.land/facebook-ad-costs-jump-21-as-lead-campaigns-struggle-while-traffic-ads-thrive/). 8.54 divided by 7.72 is 1.1062, a rise of 10.6%, derived on this page. That matches the summary, so the fall looks like the cost per lead figure repeated. Cite the rise. ## What Meta itself reports {#price} Meta's own numbers point the other way. In its [second quarter 2026 results](https://investor.atmeta.com/investor-news/press-release-details/2026/Meta-Reports-Second-Quarter-2026-Results/default.aspx), released 29 July 2026 for the quarter to 30 June 2026 and read on 7 October 2026, Meta reported average price per ad up 12% year on year, ad impressions up 14%, and revenue of $60.80 billion in US dollars, up 28%. These are Meta's figures about itself, from its investor release. They cover every advertiser across its Family of Apps worldwide, and the price is per ad shown, not per click. So clicks in a US sample of one vendor's customers got cheaper while Meta's average price per ad rose. Both can be true, because one is a median price per click on two objectives and the other is an average price per ad across every objective and every country. Neither is a SaaS price, and neither on its own measures whether advertising is getting dearer. The register of [which paid ads benchmark to cite](https://saasadsstudio.com/research/saas-ad-benchmark-sources/) records what each source does and does not measure. ## Method Every figure on this page was read on 7 October 2026 from the page named beside it. The WordStream by LocaliQ figures were read off the publisher's own 2026 and 2025 editions and second-sourced the same day to [PPC Land](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/) and [Digital Applied](https://www.digitalapplied.com/blog/facebook-ads-benchmarks-2026-cpc-cpm-ctr-industry), which are what is linked, because this site never links a listed competitor. The sources block lists those second sources, and the primaries are named in the text with their addresses in plain text. Seven figures are derived and marked as derived where they appear: 36 category rows (22 plus 14); 1,829 campaigns (1,377 plus 452); 31% (35.87 divided by 27.39); 20% overlap (3 months divided by 15); $21.08 per lead (1.80 divided by 0.0854); 30% (27.39 divided by 21.0773); and a 10.6% rise in conversion rate (8.54 divided by 7.72). Every input is a published figure from the [2026 edition](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/) or the [2025 edition](https://ppc.land/facebook-ad-costs-jump-21-as-lead-campaigns-struggle-while-traffic-ads-thrive/). No currency conversion is applied anywhere. Two sets of figures are a company describing itself and are labelled where they appear: the benchmark, which is a vendor's own customer campaigns, and Meta's price per ad, impressions and revenue, which are Meta's own reported metrics. What is deliberately absent is any first-party figure. The keyword study published on this site is Google search data and says nothing about Meta, and no signed quarterly figures file exists today, so there is no SaaS Meta number from accounts Atlas works in. ## If you are running SaaS ads on Meta The practical reading is narrow: do not grade a B2B SaaS Meta account against a median drawn from dentists, restaurants and furniture shops, and do not build a forecast by multiplying medians. Use your own cost per trial or demo, by audience and by creative, as the benchmark. If you want a structured starting point, Atlas builds your [Facebook and Instagram campaigns for SaaS](https://saasadsstudio.com/facebook-ads-for-saas/) and gives you weekly checks you run yourself, and you review and approve before anything launches, inside [the app](https://app.saasadsstudio.com). ## Questions people ask about this **What is a good Facebook ads cost per lead for a B2B SaaS in 2026?** No published benchmark answers that directly, because the most-cited set has no software row. Its nearest row, Industrial & Commercial, shows a median cost per lead of $35.87 in US dollars, but it rests on a leads sample of 452 campaigns with as few as four per category, and it counts form fills rather than trials or demos. Treat it as loose context and grade an account against its own trend. **Why do so many pages still quote $0.70 per click and $27.66 per lead?** Those are the 2025 edition's traffic cost per click and cost per lead, in US dollars. The 2026 edition replaced them on 14 September 2026 with $0.60 and $27.39, but older figures keep circulating for weeks after a new edition ships, so check the edition before citing. **Did the Facebook leads conversion rate rise or fall in the 2026 edition?** It rose. The 2025 edition published 7.72% and the 2026 edition publishes 8.54%, which is a rise of about 10.6%. One section of the 2026 report says it fell 0.98%, but that is the same figure the report gives for cost per lead, and the two editions' own numbers contradict it. **Does Meta's 12% rise in average price per ad mean Meta ads cost 12% more?** Not for any one advertiser. It is Meta's own average across every advertiser, objective and country, priced per ad shown rather than per click or per lead. In the same period the US benchmark sample recorded cheaper clicks, and both can be true because they measure different things. ## Sources 1. [PPC Land, write-up of the WordStream by LocaliQ Facebook Ads Benchmarks 2026 (published 14 September 2026)](https://ppc.land/facebook-lead-costs-hold-at-27-39-as-traffic-clicks-drop-to-0-60/), read 2026-10-07 2. [Digital Applied, Facebook Ads Benchmarks 2026 by industry, carrying the WordStream by LocaliQ 2026 tables (updated 22 September 2026)](https://www.digitalapplied.com/blog/facebook-ads-benchmarks-2026-cpc-cpm-ctr-industry), read 2026-10-07 3. [PPC Land, write-up of the WordStream by LocaliQ Facebook Ads Benchmarks 2025 (published 9 September 2025)](https://ppc.land/facebook-ad-costs-jump-21-as-lead-campaigns-struggle-while-traffic-ads-thrive/), read 2026-10-07 4. [Meta Platforms, Second Quarter 2026 Results (29 July 2026)](https://investor.atmeta.com/investor-news/press-release-details/2026/Meta-Reports-Second-Quarter-2026-Results/default.aspx), read 2026-10-07 5. [SaaS Ads Studio, Why Google Ads Benchmarks Miss SaaS](https://saasadsstudio.com/research/why-google-ads-benchmarks-miss-saas/), read 2026-10-07 ## Cite this page Free to quote with a link back. Figures last checked 2026-10-07. > The SaaS Ads Studio, "Meta Ads Benchmarks for B2B SaaS", 2026. https://saasadsstudio.com/research/meta-ads-benchmarks-for-b2b-saas/