---
title: "B2B customer acquisition cost benchmarks: what a customer actually costs by company size"
url: https://metadata.io/resources/blog/6-ways-to-lower-your-customer-acquisition-cost-cac-in-2024
description: "B2B customer acquisition cost benchmarks from $57.6M of 2025 ad spend across 153 advertisers, broken down by company size, channel and audience type."
source: metadata.io
---

Most published B2B CAC advice is written without a dataset behind it. These numbers come from 153 B2B advertisers and $57.6M of 2025 paid media spend, measured to closed-won revenue rather than to form fills. The headline finding is counterintuitive: the mid-market customer everyone chases is the most expensive one to buy.

## What is the average B2B customer acquisition cost?

Across 153 B2B advertisers running paid media in 2025, the honest blended customer acquisition cost is $58,887. Honest means every dollar of spend is counted against every won deal, including the campaigns that produced nothing, not just the campaigns that worked.

That last point is where most published CAC figures go wrong. If you divide spend by customers only for the campaigns that converted, you get a number that flatters the channel and cannot be budgeted against. The figure above does not do that.

## What is a good CAC for B2B SaaS?

There is no single good number, because CAC moves by a factor of nearly four depending on who you target. In this dataset it ranges from $35,288 for small companies on LinkedIn to $130,468 for the 501 to 1,000 employee segment. Company size, not channel, is the dominant variable.

## How does B2B CAC change by company size?

Chasing larger accounts costs more per customer, and the curve is not smooth. The 501 to 1,000 employee band is the most expensive segment in the dataset at $130,468 per customer, which is more than double the blended average and nearly four times the small-company figure.

| Target company size | Customer acquisition cost | vs blended average |
|---|---|---|
| 51–200 employees, LinkedIn | $35,288 | 40% below |
| 1,001–5,000 employees | $80,109 | 36% above |
| 501–1,000 employees | $130,468 | 122% above |
| **Blended, all advertisers** | **$58,887** | — |

The practical read: if your ICP sits in the 501 to 1,000 band, your CAC is structurally higher than the benchmark and comparing yourself to a blended average will make a healthy programme look broken. Compare against your own segment.

The full segment-by-segment cut, including the publication gates and concentration caps behind these numbers, is on the [CAC cliff finding page](https://metadata.io/benchmark-report-2026/insights/cac-cliff).

## How do you lower customer acquisition cost in B2B advertising?

Audience construction is the lever, but a smaller one than segment choice. Across 153 advertisers in 2025, prospecting against a built audience produced customers at $60,896, against $69,705 using native platform targeting: a 13% improvement on the same channels.

That is real, and it is an order of magnitude smaller than the swing between company-size segments shown above. Fix the segment question first.

| Cut | Customer acquisition cost |
|---|---|
| Prospecting, built audience | $60,896 |
| Prospecting, native platform targeting | $69,705 |
| Image ads, all channels | $59,488 |
| LinkedIn, all campaign goals | $63,312 |

Three things follow from the table, in the order they are worth doing:

1. **Pick the segment before anything else.** Moving target company size swings CAC by nearly 4x in this dataset. No amount of audience or creative work closes a gap that size.
2. **Then fix the audience.** A built audience beat native targeting by 13% on identical channels, which is worth having once the segment question is settled.
3. **Measure to won revenue, not to leads.** Cost per lead and CAC rank channels differently, which is why cheap-lead channels routinely fail to show up in closed-won.

## How were these numbers calculated?

They come from 153 B2B advertisers running paid media through Metadata during 2025, totalling $57.6M in spend, joined to CRM closed-won outcomes. Segments are published only where they clear minimum advertiser counts and concentration caps, so a single large spender cannot move a benchmark.

The full method, including what is excluded and why, is documented on the [benchmark methodology page](https://metadata.io/benchmark/methodology). Channel-level CPL, CPC, CTR and CPM for the same dataset are on [B2B advertising benchmarks](https://metadata.io/b2b-advertising-benchmarks).

## Where do your numbers land against this?

The benchmark set is ungated and filterable by company size, industry, channel, ad type and campaign goal. If your CAC sits above your segment rather than above the blended average, that is the gap worth working on.

[Open the 2026 benchmark report](https://metadata.io/benchmark-report-2026).