Wasted B2B Ad Spend on Low-Intent Audiences: 2026 Benchmark Study
Wasted B2B Ad Spend on Low-Intent Audiences: 2026 Benchmark Study
By Dale Brett, Founder & CEO
Published B2B research — most notably the Demandbase and eMarketer "From Ad Waste to ROI" study (April 2025) — finds that 58% of B2B marketers consider low-intent audience waste a significant problem, and that more than half estimate a meaningful share of their spend is being wasted on the wrong accounts. The operational fix is not better creative. It is better pre-targeting — specifically, filtering ad audiences through intent signals before the impression is served. This report compiles the published benchmarks, explains the structural causes, and lays out the intent-based pre-filter approach that most reduces waste in practice.
Why this report exists
Every B2B marketing leader in 2026 is being asked to justify ad spend against pipeline — not clicks, not MQLs, pipeline. The CFO conversation has changed. Published research is loud about the size of the waste problem, but the category of advice on how to fix it is dominated by vendors pitching their own tool. We wanted a neutral compilation of what the published numbers actually say, what mechanism causes the waste, and what the working fix looks like.
FL0 is an intent signals engine that runs AI go-to-market agents to win you new accounts. We sell into exactly this problem space — the gap between "who is actually in-market" and "who is actually being reached" — and the patterns in this report draw on conversations with demand-gen and RevOps leaders evaluating how to stop paying to reach the wrong audiences.
The published waste figures
The cleanest primary source on B2B ad-spend waste with a named author, named publisher, and a published date is the Demandbase + eMarketer study:
Demandbase — From Ad Waste to ROI (April 2025) — the study reports that 58% of B2B marketers describe low-intent audience waste as a significant issue, and that more than half estimate a meaningful portion of their spend is reaching the wrong accounts. Figures in this study are sourced to eMarketer's underlying dataset.
Other roundups in the category cite higher aggregate percentages (such as 64% of budget "wasted on irrelevant keywords" or 80% of impressions reaching out-of-market buyers), but we could not trace those numbers back to a primary dataset with published methodology. We are excluding them rather than reprinting them.
Methodology
This report synthesizes three input categories. First, published benchmark reports that link back to a named dataset with a published date — the Demandbase + eMarketer study is the clearest of these. Second, academic and browser-vendor documentation on IP-based targeting accuracy and third-party cookie attrition — the structural substrate that determines how much waste is inevitable regardless of strategy. Third, qualitative conversations with demand-gen and revenue-operations practitioners about how audience construction actually works in production, presented as practitioner observation rather than measured data.
No vendor sponsored this report.
Structural causes — why the waste is not a creative problem
A reasonable first reaction to "a significant share of B2B ad spend is wasted" is to suspect creative quality. The published substrate data says otherwise. Three structural forces drive most of the waste, and none of them are fixed by better copy or design.
Cause 1 — IP-based targeting accuracy has collapsed. B2B ad targeting historically relied on mapping an impression's source IP to a company identity. Remote work since 2020 has broken that assumption at scale. Academic work on IP-based geolocation and company identification documented significant accuracy limits even under ideal conditions (IEEE paper on IP geolocation). Post-remote-work, the real-world figure is materially lower, because employees are on home, coffee-shop, and consumer-ISP networks that map to nothing.
Cause 2 — Third-party cookies are dying. Apple's Intelligent Tracking Prevention blocks third-party cookies by default in Safari (Apple WebKit documentation). Firefox's Enhanced Tracking Protection does the same (Mozilla support documentation). Google Chrome moved against third-party cookies through its Privacy Sandbox program (Google Privacy Sandbox timeline). The substrate that most B2B ad-targeting products were built on no longer works across the majority of modern browser traffic. The tools still ship targeting features, but the inputs have thinned substantially.
Cause 3 — Intent is inferred too late in the funnel. Most B2B ad platforms infer intent from on-site behavior after the click. That means you have already paid for the impression and the click before you know whether the prospect is actually in-market. Retargeting amplifies the waste instead of correcting it — you pay twice to reach the same out-of-market person, once on the first touch, again on the retarget.
The common thread: the cost of reaching a low-intent audience has gone up, and the ability to exclude them has gone down. Waste is the predictable output of that scissors motion. Creative will not fix it.
The intent pre-filter approach
The intervention that most consistently reduces waste in the published case studies, and that matches what revenue operators describe doing in practice, is audience pre-filtering through intent data before the impression is served. The mechanism is straightforward:
Identify accounts showing intent signals — first-party website activity, product telemetry, real-time signal detection across the open web, third-party surge data, or a blend — that indicate they are actively researching your category.
Upload that account list to your ad platform as a matched audience (LinkedIn Matched Audiences, Google Customer Match, Meta Custom Audiences, or direct integration via a DSP).
Restrict campaign delivery to that list. Do not layer intent filtering on top of broad targeting — replace the broad targeting.
Refresh the intent audience on a weekly or daily cadence so it reflects current surge rather than stale signal.
The effect is to shift the impression budget from a broadly-targeted population (where a meaningful share is structurally wasted) to a tightly-filtered population (where every impression is at least correlated with in-market behavior). Waste cannot go to zero because intent data itself has error rates, but the practical reduction is substantial.
What the working stack looks like
A 2026 B2B ad-spend stack optimized against waste runs four layers in sequence:
Layer 1 — First-party intent capture. Website analytics, product telemetry, form fills, pricing-page visits. Real-time. Lowest latency, highest accuracy. This is the primary signal. Every dollar of ad-platform budget should be filtered through an audience derived from this layer first.
Layer 2 — Intent signals engine + GTM agents. The gap in most ad-waste stacks is that pure data tools (Bombora, 6sense, Clearbit) surface signals without acting on them, and pure outbound tools (Apollo, Instantly) act without surfacing signals. FL0 is an intent signals engine that runs AI go-to-market agents to win you new accounts — it sits in the gap by combining the signal layer (a global intent data graph) with agents that engage in-market accounts directly, and by feeding the resulting account list into whatever matched-audience builder the team already runs. The practical effect on ad waste is that impression spend is restricted to accounts that a real-time intent source has already flagged as in-market, instead of to a broad firmographic population.
Layer 3 — De-anonymization. Tools that market the ability to connect anonymous website traffic to a company identity (Clearbit Reveal, RB2B, Leadfeeder, and others — see each vendor's product page for specifics). This expands Layer 1 without introducing cooperative-data noise.
Layer 4 — Third-party surge overlay. Bombora, 6sense, or Demandbase used not as the primary signal but as a prioritization lens over the earlier layers. Used this way, the category's structural weaknesses (latency, false positives, multi-divisional noise) are mostly absorbed.
Layer 5 — Ad-platform audience sync. The combined audience flows into LinkedIn Matched Audiences, Google Customer Match, Meta Custom Audiences, and direct DSP integration — refreshed at least weekly, ideally daily. This is where the actual impression spend lands.
The common failure mode is to skip Layers 1 and 2 and run only Layer 4. That is effectively paying a third-party surge vendor to target your broad audience slightly less broadly, which is not the same as filtering for in-market intent. The published waste figures are consistent with a category where most teams run exactly that setup.
Limitations
Only one cleanly sourced primary dataset. The Demandbase + eMarketer study is the only waste figure in this report that traces back to a named author, publisher, and date. Roundups citing specific percentages like 64% or 80% are widely circulated but we could not verify their methodology, so we have not reprinted them.
No controlled experiment. A rigorous test would require randomized treatment and control at the campaign level across multiple advertisers, with pipeline-attribution methodology. No such experiment exists in the public literature as of April 2026. The category is ripe for one.
The waste problem is scale-dependent. Published benchmarks mix enterprise and mid-market advertisers. Enterprise advertisers with dedicated data-ops teams can drive waste below the published median; a seed-stage founder running a small LinkedIn budget will likely sit above it. Benchmark your own number; do not assume the average applies to you.
Vendor capabilities cited in this report are vendor-stated. Any description of what a named ad platform or intent tool does is drawn from the vendor's own product page, not from independent testing. Verify before purchase.
A 30-day reduction roadmap
The working fix for most B2B advertisers is a 30-day rebuild of the audience construction layer, not a creative refresh. The sequence below compresses what most teams iterate over six months into a structured four-week program.
Week 1 — baseline your current waste rate. Pull the last 90 days of ad-platform data across LinkedIn, Google, and Meta. Join to CRM. Calculate the percentage of clicks that resulted in a sales-accepted opportunity. Whatever that number is, subtract it from 100 — that is your current waste rate. Most teams doing this exercise for the first time find they have never explicitly measured the number.
Week 2 — stand up first-party capture. Confirm GA4, a form-capture tool, and product telemetry (if applicable) are writing into a source of truth your ad-platform audience builder can read. This is the layer most teams under-invest in because it is not glamorous. It is also the highest-leverage single change in the whole roadmap.
Week 3 — add de-anonymization and build a matched audience. Plug in a de-anonymization tool. Build a matched audience from the combined first-party and de-anonymized account list. Upload to LinkedIn Matched Audiences, Google Customer Match, and Meta Custom Audiences. Set campaigns to deliver only to that list.
Week 4 — measure the delta. Compare cost per qualified account for the week-4 campaigns against the week-1 baseline. The delta varies widely depending on how broad the prior baseline was. Teams coming from pure firmographic targeting typically see the largest improvements; teams already running some form of intent layer see smaller ones.
Week | Action | Expected deliverable |
|---|---|---|
1 | Measure baseline waste rate | Documented percentage |
2 | Stand up first-party capture layer | Source of truth wired to ad builder |
3 | Add de-anonymization, build matched audience, restrict delivery | Campaigns live against intent audience |
4 | Measure cost per qualified account delta | Signed-off report vs baseline |
Nothing in the 30-day plan requires buying an enterprise intent-data contract. The most common mistake in ad-waste reduction programs is to skip Weeks 1 and 2 and spend the budget on a third-party surge subscription in Week 3. That reverses the priority order and delivers the smallest reduction per dollar spent.
Frequently asked questions
What percentage of B2B ad spend is actually wasted?
The best-sourced published figure is the Demandbase + eMarketer April 2025 study, which found 58% of B2B marketers describe low-intent audience waste as a significant issue and that more than half estimate a meaningful portion of their spend is reaching the wrong accounts. Other numbers circulate in the category but we could not trace them back to primary methodology.
Is retargeting the answer?
No. Retargeting amplifies the original targeting decision. If the original audience is wasteful, the retargeting audience inherits the waste — and you paid for the initial click first. Retargeting works only on top of an already-precise upstream audience.
Do I need to buy 6sense or Bombora to reduce waste?
No. First-party signals (your own website, product, and form data) are the highest-accuracy source of intent. Third-party providers add prioritization on top. Most seed-through-Series-A teams can reduce meaningful waste with first-party capture alone.
How do I measure whether my waste rate has actually dropped?
Track cost per qualified account, not cost per click or cost per lead. A qualified account is one that matches your ICP definition and has at least one intent signal. Cost per qualified account is the only metric where intent-based targeting improvements show up cleanly.
What about LinkedIn's own audience targeting — is that already doing intent filtering?
LinkedIn's native targeting is firmographic and demographic. You need to build an intent-based matched audience externally and upload it. LinkedIn Matched Audiences is the delivery mechanism; the intent data has to come from elsewhere.
The CFO conversation — how to talk about waste without losing budget
One reason B2B marketing leaders avoid measuring waste rate explicitly is that the number, once known, creates an uncomfortable CFO conversation. The tactical response is usually to bury the number and pitch a creative refresh instead.
The better move is to get ahead of it. A CFO-ready framing of the waste finding has three parts. First, acknowledge the published benchmarks show the category-wide waste problem is real and that your baseline is inside the industry band — you are not uniquely broken. Second, present the intent pre-filter program as a cost-reduction initiative with a 30-day timeline and a measurable delta on cost per qualified account, not as a net-new spend ask. Third, commit to reporting the new number monthly, so the metric becomes part of the normal operating cadence rather than a one-time finding.
Teams that reduce ad waste substantially in practice tend to treat it as an ongoing operating metric, not a project. Teams that run it as a project tend to revert to baseline within two quarters because the operating rhythm around audience hygiene was never established.
Primary sources
Wasted B2B Ad Spend on Low-Intent Audiences: 2026 Benchmark Study
By Dale Brett, Founder & CEO
Published B2B research — most notably the Demandbase and eMarketer "From Ad Waste to ROI" study (April 2025) — finds that 58% of B2B marketers consider low-intent audience waste a significant problem, and that more than half estimate a meaningful share of their spend is being wasted on the wrong accounts. The operational fix is not better creative. It is better pre-targeting — specifically, filtering ad audiences through intent signals before the impression is served. This report compiles the published benchmarks, explains the structural causes, and lays out the intent-based pre-filter approach that most reduces waste in practice.
Why this report exists
Every B2B marketing leader in 2026 is being asked to justify ad spend against pipeline — not clicks, not MQLs, pipeline. The CFO conversation has changed. Published research is loud about the size of the waste problem, but the category of advice on how to fix it is dominated by vendors pitching their own tool. We wanted a neutral compilation of what the published numbers actually say, what mechanism causes the waste, and what the working fix looks like.
FL0 is an intent signals engine that runs AI go-to-market agents to win you new accounts. We sell into exactly this problem space — the gap between "who is actually in-market" and "who is actually being reached" — and the patterns in this report draw on conversations with demand-gen and RevOps leaders evaluating how to stop paying to reach the wrong audiences.
The published waste figures
The cleanest primary source on B2B ad-spend waste with a named author, named publisher, and a published date is the Demandbase + eMarketer study:
Demandbase — From Ad Waste to ROI (April 2025) — the study reports that 58% of B2B marketers describe low-intent audience waste as a significant issue, and that more than half estimate a meaningful portion of their spend is reaching the wrong accounts. Figures in this study are sourced to eMarketer's underlying dataset.
Other roundups in the category cite higher aggregate percentages (such as 64% of budget "wasted on irrelevant keywords" or 80% of impressions reaching out-of-market buyers), but we could not trace those numbers back to a primary dataset with published methodology. We are excluding them rather than reprinting them.
Methodology
This report synthesizes three input categories. First, published benchmark reports that link back to a named dataset with a published date — the Demandbase + eMarketer study is the clearest of these. Second, academic and browser-vendor documentation on IP-based targeting accuracy and third-party cookie attrition — the structural substrate that determines how much waste is inevitable regardless of strategy. Third, qualitative conversations with demand-gen and revenue-operations practitioners about how audience construction actually works in production, presented as practitioner observation rather than measured data.
No vendor sponsored this report.
Structural causes — why the waste is not a creative problem
A reasonable first reaction to "a significant share of B2B ad spend is wasted" is to suspect creative quality. The published substrate data says otherwise. Three structural forces drive most of the waste, and none of them are fixed by better copy or design.
Cause 1 — IP-based targeting accuracy has collapsed. B2B ad targeting historically relied on mapping an impression's source IP to a company identity. Remote work since 2020 has broken that assumption at scale. Academic work on IP-based geolocation and company identification documented significant accuracy limits even under ideal conditions (IEEE paper on IP geolocation). Post-remote-work, the real-world figure is materially lower, because employees are on home, coffee-shop, and consumer-ISP networks that map to nothing.
Cause 2 — Third-party cookies are dying. Apple's Intelligent Tracking Prevention blocks third-party cookies by default in Safari (Apple WebKit documentation). Firefox's Enhanced Tracking Protection does the same (Mozilla support documentation). Google Chrome moved against third-party cookies through its Privacy Sandbox program (Google Privacy Sandbox timeline). The substrate that most B2B ad-targeting products were built on no longer works across the majority of modern browser traffic. The tools still ship targeting features, but the inputs have thinned substantially.
Cause 3 — Intent is inferred too late in the funnel. Most B2B ad platforms infer intent from on-site behavior after the click. That means you have already paid for the impression and the click before you know whether the prospect is actually in-market. Retargeting amplifies the waste instead of correcting it — you pay twice to reach the same out-of-market person, once on the first touch, again on the retarget.
The common thread: the cost of reaching a low-intent audience has gone up, and the ability to exclude them has gone down. Waste is the predictable output of that scissors motion. Creative will not fix it.
The intent pre-filter approach
The intervention that most consistently reduces waste in the published case studies, and that matches what revenue operators describe doing in practice, is audience pre-filtering through intent data before the impression is served. The mechanism is straightforward:
Identify accounts showing intent signals — first-party website activity, product telemetry, real-time signal detection across the open web, third-party surge data, or a blend — that indicate they are actively researching your category.
Upload that account list to your ad platform as a matched audience (LinkedIn Matched Audiences, Google Customer Match, Meta Custom Audiences, or direct integration via a DSP).
Restrict campaign delivery to that list. Do not layer intent filtering on top of broad targeting — replace the broad targeting.
Refresh the intent audience on a weekly or daily cadence so it reflects current surge rather than stale signal.
The effect is to shift the impression budget from a broadly-targeted population (where a meaningful share is structurally wasted) to a tightly-filtered population (where every impression is at least correlated with in-market behavior). Waste cannot go to zero because intent data itself has error rates, but the practical reduction is substantial.
What the working stack looks like
A 2026 B2B ad-spend stack optimized against waste runs four layers in sequence:
Layer 1 — First-party intent capture. Website analytics, product telemetry, form fills, pricing-page visits. Real-time. Lowest latency, highest accuracy. This is the primary signal. Every dollar of ad-platform budget should be filtered through an audience derived from this layer first.
Layer 2 — Intent signals engine + GTM agents. The gap in most ad-waste stacks is that pure data tools (Bombora, 6sense, Clearbit) surface signals without acting on them, and pure outbound tools (Apollo, Instantly) act without surfacing signals. FL0 is an intent signals engine that runs AI go-to-market agents to win you new accounts — it sits in the gap by combining the signal layer (a global intent data graph) with agents that engage in-market accounts directly, and by feeding the resulting account list into whatever matched-audience builder the team already runs. The practical effect on ad waste is that impression spend is restricted to accounts that a real-time intent source has already flagged as in-market, instead of to a broad firmographic population.
Layer 3 — De-anonymization. Tools that market the ability to connect anonymous website traffic to a company identity (Clearbit Reveal, RB2B, Leadfeeder, and others — see each vendor's product page for specifics). This expands Layer 1 without introducing cooperative-data noise.
Layer 4 — Third-party surge overlay. Bombora, 6sense, or Demandbase used not as the primary signal but as a prioritization lens over the earlier layers. Used this way, the category's structural weaknesses (latency, false positives, multi-divisional noise) are mostly absorbed.
Layer 5 — Ad-platform audience sync. The combined audience flows into LinkedIn Matched Audiences, Google Customer Match, Meta Custom Audiences, and direct DSP integration — refreshed at least weekly, ideally daily. This is where the actual impression spend lands.
The common failure mode is to skip Layers 1 and 2 and run only Layer 4. That is effectively paying a third-party surge vendor to target your broad audience slightly less broadly, which is not the same as filtering for in-market intent. The published waste figures are consistent with a category where most teams run exactly that setup.
Limitations
Only one cleanly sourced primary dataset. The Demandbase + eMarketer study is the only waste figure in this report that traces back to a named author, publisher, and date. Roundups citing specific percentages like 64% or 80% are widely circulated but we could not verify their methodology, so we have not reprinted them.
No controlled experiment. A rigorous test would require randomized treatment and control at the campaign level across multiple advertisers, with pipeline-attribution methodology. No such experiment exists in the public literature as of April 2026. The category is ripe for one.
The waste problem is scale-dependent. Published benchmarks mix enterprise and mid-market advertisers. Enterprise advertisers with dedicated data-ops teams can drive waste below the published median; a seed-stage founder running a small LinkedIn budget will likely sit above it. Benchmark your own number; do not assume the average applies to you.
Vendor capabilities cited in this report are vendor-stated. Any description of what a named ad platform or intent tool does is drawn from the vendor's own product page, not from independent testing. Verify before purchase.
A 30-day reduction roadmap
The working fix for most B2B advertisers is a 30-day rebuild of the audience construction layer, not a creative refresh. The sequence below compresses what most teams iterate over six months into a structured four-week program.
Week 1 — baseline your current waste rate. Pull the last 90 days of ad-platform data across LinkedIn, Google, and Meta. Join to CRM. Calculate the percentage of clicks that resulted in a sales-accepted opportunity. Whatever that number is, subtract it from 100 — that is your current waste rate. Most teams doing this exercise for the first time find they have never explicitly measured the number.
Week 2 — stand up first-party capture. Confirm GA4, a form-capture tool, and product telemetry (if applicable) are writing into a source of truth your ad-platform audience builder can read. This is the layer most teams under-invest in because it is not glamorous. It is also the highest-leverage single change in the whole roadmap.
Week 3 — add de-anonymization and build a matched audience. Plug in a de-anonymization tool. Build a matched audience from the combined first-party and de-anonymized account list. Upload to LinkedIn Matched Audiences, Google Customer Match, and Meta Custom Audiences. Set campaigns to deliver only to that list.
Week 4 — measure the delta. Compare cost per qualified account for the week-4 campaigns against the week-1 baseline. The delta varies widely depending on how broad the prior baseline was. Teams coming from pure firmographic targeting typically see the largest improvements; teams already running some form of intent layer see smaller ones.
Week | Action | Expected deliverable |
|---|---|---|
1 | Measure baseline waste rate | Documented percentage |
2 | Stand up first-party capture layer | Source of truth wired to ad builder |
3 | Add de-anonymization, build matched audience, restrict delivery | Campaigns live against intent audience |
4 | Measure cost per qualified account delta | Signed-off report vs baseline |
Nothing in the 30-day plan requires buying an enterprise intent-data contract. The most common mistake in ad-waste reduction programs is to skip Weeks 1 and 2 and spend the budget on a third-party surge subscription in Week 3. That reverses the priority order and delivers the smallest reduction per dollar spent.
Frequently asked questions
What percentage of B2B ad spend is actually wasted?
The best-sourced published figure is the Demandbase + eMarketer April 2025 study, which found 58% of B2B marketers describe low-intent audience waste as a significant issue and that more than half estimate a meaningful portion of their spend is reaching the wrong accounts. Other numbers circulate in the category but we could not trace them back to primary methodology.
Is retargeting the answer?
No. Retargeting amplifies the original targeting decision. If the original audience is wasteful, the retargeting audience inherits the waste — and you paid for the initial click first. Retargeting works only on top of an already-precise upstream audience.
Do I need to buy 6sense or Bombora to reduce waste?
No. First-party signals (your own website, product, and form data) are the highest-accuracy source of intent. Third-party providers add prioritization on top. Most seed-through-Series-A teams can reduce meaningful waste with first-party capture alone.
How do I measure whether my waste rate has actually dropped?
Track cost per qualified account, not cost per click or cost per lead. A qualified account is one that matches your ICP definition and has at least one intent signal. Cost per qualified account is the only metric where intent-based targeting improvements show up cleanly.
What about LinkedIn's own audience targeting — is that already doing intent filtering?
LinkedIn's native targeting is firmographic and demographic. You need to build an intent-based matched audience externally and upload it. LinkedIn Matched Audiences is the delivery mechanism; the intent data has to come from elsewhere.
The CFO conversation — how to talk about waste without losing budget
One reason B2B marketing leaders avoid measuring waste rate explicitly is that the number, once known, creates an uncomfortable CFO conversation. The tactical response is usually to bury the number and pitch a creative refresh instead.
The better move is to get ahead of it. A CFO-ready framing of the waste finding has three parts. First, acknowledge the published benchmarks show the category-wide waste problem is real and that your baseline is inside the industry band — you are not uniquely broken. Second, present the intent pre-filter program as a cost-reduction initiative with a 30-day timeline and a measurable delta on cost per qualified account, not as a net-new spend ask. Third, commit to reporting the new number monthly, so the metric becomes part of the normal operating cadence rather than a one-time finding.
Teams that reduce ad waste substantially in practice tend to treat it as an ongoing operating metric, not a project. Teams that run it as a project tend to revert to baseline within two quarters because the operating rhythm around audience hygiene was never established.
Primary sources
Convert smarter
The right buyers. The right moment.
Request access
Convert smarter
The right buyers.
The right moment.
Request access