Cost per SQL is your total program spend divided by the number of sales qualified leads it produced, and the published 2026 benchmarks disagree wildly: SaaS Hero calls under $400 strong performance for paid B2B SaaS programs, a widely cited First Page Sage benchmark report puts the blended B2B average near $1,357, and Two Spouts reports a median around $3,500 for cybersecurity SQLs from Google Ads. That spread is not measurement noise. It is mostly definition and channel, and this article breaks down what a sales qualified lead should cost by channel, why outbound SQLs price on a different curve than paid-ads SQLs, and how to compute a number for your own program that you can defend.
We spend a lot of time on this metric because we sit on the supply side of it. Our parent outbound agency, Referral Program Pros, has run more than 4,000 outbound campaigns and booked over 7,000 meetings, and the qualified-lead math behind those campaigns is the playbook GTM Bud was built on. The platform backs the top of that funnel with a written floor: a 5 percent positive reply rate on LinkedIn and 1.5 percent on email, or a full refund. Where those figures appear in this article, they are guaranteed minimums, labeled as such.
Sources: every external number here is attributed to a named publisher: First Page Sage, SaaS Hero, Two Spouts, GrowthSpree, LeadHaste citing Sopro, Understory, Belkins, Ebsta and Pavilion, Operatix, Martal, RepVue, MIT (Hadzima), DanishLeadCo, OutboundSalesPro, and Bridge Group data via SDR benchmark roundups. Worked examples are arithmetic from labeled assumptions.
What is a good cost per SQL in 2026?
A good cost per SQL for B2B in 2026 is roughly $200 to $500 for efficient paid-acquisition SaaS programs, with SaaS Hero’s 2026 benchmarks calling anything under $400 strong when at least 20 percent of SQLs convert to opportunities. Blended across all channels and industries, the widely cited First Page Sage benchmark report puts the B2B average near $1,357, and GrowthSpree’s quality-adjusted 2026 data spans $200 to $500 in B2B manufacturing up to $800 to $2,000 in cybersecurity. The honest answer is that a universal target does not exist, because deal economics set the ceiling: a $1,500 SQL is healthy against a $40,000 contract and fatal against a $4,000 one. Judge your number against your own deal size and channel mix, and only compare against benchmarks that define a sales qualified lead the way your team does.
The rest of this article shows where those ranges come from, starting with the definition problem that quietly breaks most comparisons.
The MQL versus SQL trap
A sales qualified lead is a lead the sales team has vetted and accepted as a genuine opportunity, typically against criteria like budget, authority, need, and timeline. A marketing qualified lead is a lead that cleared a marketing-defined engagement bar, such as downloading a whitepaper or visiting a pricing page, with no human confirmation that a real buying process exists. Those are different funnel stages separated by a brutal conversion step: Understory’s 2026 B2B SaaS data puts the baseline MQL-to-SQL rate near 13 percent, while SaaS Hero’s 2026 benchmarks report an 18 to 22 percent average for B2B SaaS, with top performers reaching 25 to 35 percent. Any published benchmark that does not say which of the two stages it prices is not really a benchmark; it is a marketing number waiting to be misread.
Here is why that matters for benchmarking. At those conversion rates, an SQL costs five to eight times what an MQL costs from the same channel, before any change in spend. A vendor quoting $180 “qualified leads” that are really MQLs is more expensive than a vendor quoting $900 SQLs, once you divide through the conversion step. The same trap runs one level lower: Belkins’ 2026 benchmark data puts the average B2B SaaS cost per raw lead near $237, and LeadHaste’s 2026 qualified-lead benchmarks note raw cost per lead spanning roughly $25 for referrals to $840 for trade shows, per Sopro data. None of those numbers say anything about SQLs until you multiply through your funnel.
So before you benchmark anything, write down which stage you are pricing. Every figure in this article is a sales-accepted, vetted lead unless stated otherwise, and your own tracking should hold the same line. The downstream stages, from SQL to revenue, are covered in our outbound ROI measurement guide; this article stays on the acquisition side of the metric.
Cost per SQL benchmarks by channel and vertical
Here are the published 2026 figures side by side. The sources disagree at the edges, sometimes by multiples, so treat overlaps as the trustworthy zone and every single-source number as that source’s claim.
| Channel or segment | Published cost per SQL | Source |
|---|---|---|
| Blended B2B average, all channels | About $1,357 | First Page Sage benchmark report, widely cited |
| Efficient paid-acquisition SaaS programs | $200 to $500; under $400 is strong | SaaS Hero 2026 benchmarks |
| Google Ads, B2B SaaS median | $800 to $2,500 | GrowthSpree data via Two Spouts 2026 |
| Google Ads, DevTools | About $650 | Two Spouts 2026 vertical benchmarks |
| Google Ads, cybersecurity | About $3,500 | Two Spouts 2026 vertical benchmarks |
| Quality-adjusted, B2B manufacturing | $200 to $500 | GrowthSpree 2026 quality-adjusted benchmarks |
| Quality-adjusted, vertical SaaS | $400 to $900 | GrowthSpree 2026 quality-adjusted benchmarks |
| Quality-adjusted, cybersecurity | $800 to $2,000 | GrowthSpree 2026 quality-adjusted benchmarks |
| Outbound agency, derived from per-meeting | Roughly $450 to $1,200 (arithmetic, see below) | Derived from DanishLeadCo per-meeting ranges |
| In-house SDR, derived from per-meeting | Roughly $2,100 to $3,450 (arithmetic, see below) | Derived from DanishLeadCo and OutboundSalesPro |
Two notes on reading that honestly. First, cybersecurity appears three times with three different answers: SaaS Hero reports $500 to $900, GrowthSpree $800 to $2,000, and Two Spouts a $3,500 Google Ads median. That is what happens when sources mix channels and SQL definitions, and it is exactly why you should read every benchmark as a range with an asterisk rather than a target with a decimal point.
Second, the two outbound rows are not published benchmarks; they are arithmetic. Published outbound benchmarks price meetings, not SQLs, so we converted using the standard chain: SDR benchmark roundups place meeting-to-opportunity conversion between roughly 25 and 50 percent, and at the conservative one-in-three rate, cost per SQL is cost per held meeting times three. DanishLeadCo’s $150 to $400 agency meetings become $450 to $1,200 SQLs; the $700 to $1,150 fully loaded in-house SDR meetings from DanishLeadCo and OutboundSalesPro become $2,100 to $3,450 SQLs. The full per-meeting layer of this math, including a complete worked SDR example, lives in our cost per meeting benchmarks guide.
Why do outbound SQLs price differently from paid-ads SQLs?
Because the two channels buy different inputs. Paid ads buy clicks in an auction, so cost per SQL floats with competition: when cybersecurity CPCs exceed $25 to $40, as Two Spouts reports for high-intent security terms, the SQL price inflates no matter how good your funnel is. Outbound buys capacity, meaning research, sending infrastructure, and sequencing labor, at prices that are largely fixed per month. That single difference produces three practical consequences. First, outbound cost per SQL responds to targeting quality rather than auction pressure, so it improves with list discipline instead of budget. Second, an outbound SQL skips the MQL stage entirely: the lead is qualified in a live conversation, not inferred from a form fill, so the 13 to 22 percent MQL-to-SQL leak never appears in the math. Third, outbound scales stepwise with capacity added, while paid scales smoothly but at rising marginal cost per click.
Neither structure is universally cheaper, and the derived table rows above show outbound spanning both sides of the paid-ads medians depending on who does the work. The honest comparison is marginal: ask what the next ten SQLs cost from each channel at your current saturation, not what the average was last quarter. The full cost side of the outbound column, tools, data, people, and the hidden line items, is itemized in our guide to how much B2B outbound costs in 2026.
How to compute your own cost per SQL
The formula is total program cost for a period divided by SQLs accepted by sales in that same period. The discipline is in the inputs: all of the cost, including labor and management, and only genuinely accepted leads, not calendar invites. Here is the outbound version, every assumption labeled so you can swap in your own numbers.
- Program cost: take a fully loaded in-house SDR at $119,200 per year, about $9,933 per month. That is RepVue’s median $85,000 on-target earnings plus 25 percent benefits per the MIT (Hadzima) fully loaded rule, plus $7,200 in tools and data, plus $12,000 in allocated management, the same build-up worked line by line in our cost per meeting benchmarks guide.
- Held meetings: Bridge Group data via SDR benchmark roundups puts median meetings set at 14.6 per month, and Operatix benchmarks put attendance near 80 percent, so about 11.7 meetings actually happen.
- SQLs: at the conservative one-in-three meeting-to-opportunity rate, 11.7 held meetings produce about 3.9 SQLs per month.
- Cost per SQL: $9,933 divided by 3.9 is roughly $2,550 per SQL, which is why the in-house row in the table sits where it does. Run the same division on bookings instead of accepted SQLs and you get a flattering $680 figure that prices calendar invites, not qualified pipeline.
Then sanity-check the result against your deal economics rather than against the table. At Ebsta and Pavilion’s published average B2B win rate near 19 percent, roughly one SQL in five becomes revenue, so a $2,550 SQL implies about $12,750 of qualified-pipeline cost per closed deal. Against a $60,000 contract that is around 21 percent of first-year revenue; against a $10,000 contract it is unpayable. How many SQLs you need in the first place is the reverse of this math, worked stage by stage in our guide to how many leads you need to hit a revenue goal.
An explicitly illustrative platform calculation, every assumption yours to replace: one GTM Bud LinkedIn sending account costs a flat $350 per month with 1,200 leads per month included. At the guaranteed floor of 5 percent positive replies, that is 60 positive replies; at one booking per three replies, the middle of Martal’s published 25 to 40 percent range, 20 booked meetings; at Operatix’s 80 percent show benchmark, 16 held; at one SQL per three held meetings, about 5.3 SQLs. $350 divided by 5.3 is roughly $66 per SQL in platform cost. That figure is arithmetic on labeled assumptions, not a promised outcome: it excludes your time on replies and calls, and everything after the reply depends on your offer. What it does show is the structural point from the previous section: when the automated lead generation layer replaces the labor line, the labor-dominated $2,550 figure collapses, because the cost side was never really about software.
What moves the number: qualification rate, not spend
Cost per SQL has one numerator and two multiplied denominators, reply-to-meeting and meeting-to-SQL, and the second one is the lever most teams never touch. Moving meeting-to-SQL conversion from one in four to one in three cuts cost per SQL by 25 percent with zero new spend, and that conversion is almost entirely a targeting decision made weeks earlier: meetings with buyers who match your ICP qualify, meetings with the merely curious do not. This is the argument for pointing an AI outbound sales tool at a narrow, well-defined segment rather than widening the list when the SQL count disappoints, and for treating a falling qualification rate as a list problem before treating it as a sales problem.
The one lever to refuse is the definitional one. Loosening what counts as “qualified” improves every SQL metric instantly and ruins the only metric that matters, cost per closed deal, because unqualified pipeline still consumes your scarcest resource: closing time.
Frequently asked questions about cost per SQL
What is the difference between cost per MQL and cost per SQL?
Cost per MQL prices a marketing-defined signal, such as a whitepaper download or a pricing-page visit, while cost per SQL prices a lead the sales team has vetted and accepted as a genuine opportunity. Because published B2B SaaS benchmarks put MQL-to-SQL conversion between roughly 13 and 22 percent, cost per SQL typically runs five to eight times cost per MQL from the same channel. Comparing one vendor quoting MQL prices against another quoting SQL prices is the single most common way teams misjudge a channel.
What percentage of MQLs become SQLs?
Published 2026 B2B benchmarks cluster between 13 and 22 percent. Understory puts the B2B SaaS baseline near 13 percent, while SaaS Hero reports an 18 to 22 percent average with top performers reaching 25 to 35 percent. The spread is mostly definitional: a strict MQL bar converts at a higher rate but produces fewer MQLs, so judge the funnel on cost per SQL rather than on either stage alone.
Is a sales qualified lead the same as a sales opportunity?
Not always, and the gap matters for benchmarking. In most definitions an SQL is a lead sales has vetted and agreed to work, and it becomes an opportunity once discovery confirms budget, authority, need, and timeline. Some teams collapse the two stages into one. Before comparing your cost per SQL against any benchmark, check whether the source counts sales-accepted leads or confirmed opportunities, because the same funnel produces roughly two to three times more of the former.
How many SQLs does a B2B company need per month?
Work backwards from revenue. At the roughly 19 percent average B2B win rate published by Ebsta and Pavilion’s GTM benchmarks, a $500K annual target at a $10K average deal needs about 250 qualified opportunities a year, or around 21 a month. Halve the deal size and the requirement doubles. The right number is your revenue goal divided by deal size and then by win rate, not a borrowed figure, and the lead volume behind it is worked in our guide to how many leads you need to hit a revenue goal.
How do you lower cost per SQL without loosening qualification?
Improve the stages before qualification instead of redefining the bar. Tighter targeting raises reply and meeting rates from the same spend, automated lead generation removes the research and sequencing labor that dominates outbound cost, and faster follow-up recovers meetings and conversations you already paid for. Loosening the SQL definition lowers the metric while quietly raising cost per closed deal, which is the number that actually pays the bills.
Price the pipeline stage that actually predicts revenue
Cost per SQL is the first number in the funnel that prices something sales agreed is real, which makes it worth computing honestly: full program cost divided by sales-accepted leads, benchmarked only against sources that define the stage the way you do and sell at your deal size. Run the arithmetic above with your own inputs, and if the labor-dominated version of the number is what you find, the fix is usually the cost structure, not the funnel. GTM Bud runs the research, personalization, and LinkedIn and email sequencing at a flat monthly rate per connected sending account, backed by the guarantee of 5 percent positive replies on LinkedIn and 1.5 percent on email or a full refund, with a 7-day trial to test against your own cost per SQL math. See how done-for-you outbound works and find out what a qualified lead really costs you.