...

We Audited 20+ B2B Google Ads Accounts. Here’s What We Found.

The briefing
10 findings. Skim or jump.

We audited 20+ B2B Google Ads accounts spending $10K–$150K/month. The problems were nearly identical across all of them. Wrong conversion events. Unreviewed search terms. Sub-5 Quality Scores. Slow landing pages. No CRM connection. None of it is exotic — all of it is costing you pipeline.

1
Smart Bidding is learning the wrong target
Your algorithm optimizes toward whatever you tell it — including junk leads. Fix measurement before touching bids.
2
Search terms reports collecting dust
70%+ of accounts hadn't reviewed search terms in 60+ days. Budget was funding job seekers and researchers.
70%of accounts hadn't reviewed search terms in 60+ days
3
Low Quality Score doubles your CPC
Average Quality Score was 4.2 — meaning teams were paying double vs. better-matched competitors. The landing page is the fix.
4.2average Quality Score on primary commercial keywords
4
Slow pages burn 30–50% more budget
Average mobile load time was 4.8 seconds — a 90% higher bounce probability. Nobody had measured it or tied it to CAC.
4.8saverage mobile landing page load time across audited accounts
5
Campaigns built for you, not buyers
Product-named campaigns miss how buyers actually search. Restructure around intent: problem, category, comparison, brand.
6
Broad match with no negative protection
Broad match without current negative lists consistently produces the worst CPL while showing the highest conversion volume.
7
Offline conversions missing — algorithm is blind If you only read one
90% of accounts gave Smart Bidding zero revenue signal. Import opportunity created and closed-won. This is the highest-ROI fix available.
90%of accounts had no offline conversion data imported
8
Stale assets drag Quality Score down
Low-rated RSA assets suppress ad strength and Quality Score. Removing them improved both metrics within 30 days.
9
Budget follows habit, not performance
Brand campaigns were overfunded; high-intent non-brand campaigns starved. Allocation followed launch history, not pipeline data.
10
No CRM link means no real answers
Connecting Ads to CRM data produced 20–40% CPL improvement in one quarter — without new campaigns or new creative.
20–40%CPL improvement after connecting Ads to CRM within one quarter
11
What high-performers actually do differently
Top accounts had clean tracking, weekly search term reviews, 70+ PageSpeed scores, and keyword-level revenue data. Discipline, not budget.
12
Structural neglect, not incompetence
These problems are predictable in accounts set up and then managed reactively. A systematic audit surfaces all of them.

Over the past year we have audited more than 20 B2B Google Ads accounts across mid-market companies spending between $10,000 and $150,000 per month on paid search. The accounts came from different industries, different sales cycles, and different team structures. The problems were almost identical across all of them.

What follows is not a list of edge cases. These are the findings that appeared in the majority of accounts we reviewed — the issues that are quietly consuming budget, suppressing Quality Scores, and producing leads that sales teams do not want to work. If you are running Google Ads for a mid-market B2B company, the probability that at least five of these apply to your account right now is high.

Finding 1: The Wrong Conversion Event Is Being Optimized

This was the most consistent finding across every account we reviewed. Teams were optimizing Smart Bidding toward a conversion event — typically a form fill or a demo request — without verifying that the conversion being tracked actually correlated with pipeline quality. In several accounts, the primary conversion event was firing on thank-you page visits that included non-ICP traffic: competitors, job applicants, and existing customers requesting support.

When your bidding algorithm is optimizing toward a conversion event that includes people who will never become customers, it learns to find more of those people. It gets very good at the wrong thing. The campaigns look healthy — conversion volume is up, CPA is down — while the sales team is getting leads they cannot close. The fix is not a campaign change. It is a measurement change: define what a qualified conversion actually means, instrument it correctly, and give Smart Bidding a target it can actually optimize toward.

Finding 2: Search Terms Reports Had Not Been Reviewed in Months

In more than 70% of accounts we audited, the search terms report — showing the actual queries triggering ads — had not been reviewed or actioned in 60 days or more. In accounts running broad match keywords, this meant budget was being spent on queries that had no business relationship to the product being advertised.

The most common categories of wasted spend we found in search terms reports: competitor brand terms the account had not intentionally targeted, informational queries from students and researchers, job-seeking queries (“marketing manager jobs” triggering ads for marketing software), and consumer queries from individuals who would never qualify as B2B buyers. In one account, a six-figure monthly budget had accumulated thousands of irrelevant search terms over eighteen months without a single negative keyword being added. AI-powered search query analysis can process thousands of search terms at once and surface negative keyword candidates at a scale manual review cannot match — but the starting point is reviewing the report at all.

Finding 3: Quality Scores Were Below 5 on Primary Keywords

Across the accounts we reviewed, the average Quality Score on primary commercial keywords was 4.2 out of 10. Below 5 means Google considers your ad and landing page a below-average match for the query — and charges you accordingly. In competitive B2B categories where CPCs run $15–$80, the premium paid for low Quality Score is significant: an advertiser with a Quality Score of 4 can pay double the CPC of a competitor with a Quality Score of 8 for the same ad position.

The root cause in most accounts was landing page experience. Ads were sending traffic to homepage or product pages that addressed the keyword topic somewhere on the page but did not lead with it. A visitor who clicked an ad about “B2B demand generation” arrived on a page whose headline was a generic tagline with no specific relevance to demand generation. Google registered poor expected click-through rate and poor landing page relevance. Quality Score dropped. CPCs increased. The team increased budget to compensate. The landing page is the single highest-leverage variable in paid search performance — and it was the most neglected element in the majority of accounts we reviewed.

Finding 4: Landing Pages Were Loading in 4+ Seconds on Mobile

We ran every primary landing page URL through Google PageSpeed Insights on mobile. The average score across all accounts was 41 out of 100. The average load time was 4.8 seconds. Not a single account we audited had a primary landing page scoring above 70 on mobile.

At 4.8 seconds load time, Google’s research indicates the probability of a bounce is approximately 90% higher than at one second. For accounts spending $30,000 per month on paid search, this means a substantial portion of every dollar spent is going toward traffic that bounces before the page finishes loading — generating no impression of the offer, no form view, and no opportunity to convert, while still consuming the full click cost. A slow landing page increases effective CAC by 30–50% through lower Quality Score and higher bounce rates — a growth problem that in every account we reviewed, nobody had measured or connected to their campaign economics.

Finding 5: Campaigns Were Structured Around Products, Not Buyer Intent

The majority of accounts were structured around internal product categories rather than buyer intent signals. A company selling marketing analytics software had campaigns named after their product features — “Reporting Dashboard,” “Attribution Module,” “Integration Suite” — rather than the problems those features solve. The keywords in each campaign reflected how the company thought about its product, not how buyers searched for a solution.

Buyers do not search for “attribution module.” They search for “how to track marketing ROI,” “marketing attribution software for B2B,” and “which marketing channels are driving revenue.” Campaigns built around product taxonomy instead of buyer intent language systematically miss the queries that matter while capturing ones that do not. Restructuring around intent — problem-awareness, category research, vendor comparison, and brand — produces fundamentally different coverage of the search landscape and typically reduces wasted spend while increasing qualified traffic.

Finding 6: Broad Match Was Running Without Adequate Negative Keyword Protection

Google has aggressively pushed advertisers toward broad match keywords over the past two years, with the pitch that Smart Bidding can handle the increased query variance. In B2B accounts, this is significantly more dangerous than in consumer accounts — because the universe of irrelevant queries for a B2B product is enormous and the cost per click for accidentally targeting them is high.

In every account running broad match at meaningful spend, we found the same pattern: negative keyword lists that had been copied from the original campaign build and never updated. Broad match had been introduced months or years later but the negative list protecting it was built for a different campaign structure. Queries the original structure would have never triggered were now entering the broad match auction with no exclusion in place.

Broad match with Smart Bidding can work in B2B — but only with aggressive, continuously maintained negative keyword lists and conversion data that is clean enough for the algorithm to learn from. Without both conditions met, broad match in B2B accounts consistently produces the worst CPL of any match type while appearing to generate the most conversion volume.

Finding 7: No Offline Conversion Data Was Being Imported

In 90% of the accounts we audited, the only conversion events being tracked were online actions: form fills, demo requests, contact page visits. Not a single one was importing offline conversion data — opportunities created, deals closed, revenue generated — back into Google Ads.

This means Smart Bidding in every one of those accounts was optimizing toward form fills from anyone willing to fill out a form — including poor-fit leads, tire kickers, and competitors doing research. The algorithm had no signal telling it which form fills actually became pipeline and which ones were worthless. It cannot learn to find more of your ideal customers when it has no data telling it who your ideal customers are.

Importing offline conversions — at minimum, “opportunity created” and “deal closed won” — gives Smart Bidding the data it needs to optimize toward revenue rather than activity. It requires a CRM integration and some technical setup, but the performance impact is consistently the highest-ROI improvement available in a B2B Google Ads account that is already generating meaningful conversion volume. Connecting marketing spend to closed revenue is the measurement infrastructure that makes every other optimization more effective — and in the context of Google Ads, it directly improves what the algorithm is trying to do.

Finding 8: Ad Copy Had Not Been Updated in 6+ Months

Creative fatigue in paid search is less visible than in display or social but equally real. In the majority of accounts, the responsive search ad asset combinations in primary campaigns had not been refreshed in six months or more. Google’s asset performance ratings showed a significant portion of headlines and descriptions rated “Low” — meaning the algorithm had tested them, found them underperforming, and deprioritized them — but they had not been removed or replaced.

The compounding problem: low-performing assets in an RSA reduce the overall ad strength rating, which contributes to lower expected click-through rate, which feeds into lower Quality Score. Removing consistently low-performing assets and replacing them with fresh variations addressing current buyer pain points — particularly messaging relevant to current market conditions — consistently improved ad strength ratings and Quality Score within 30 days of implementation.

Finding 9: Budgets Were Allocated by Channel History, Not by Performance

In the majority of accounts, monthly budget allocation across campaigns followed a pattern established at the account’s original build — with minor adjustments made reactively when a campaign ran out of budget or dramatically underperformed. Almost none were systematically reallocating budget based on which campaigns produced the lowest CPL against the highest lead quality.

The most common consequence: brand campaigns were over-funded relative to their actual incremental value. Brand keyword CPCs are low, conversion rates are high, and CPL looks excellent — but much of that brand traffic would have come through organic search if the paid brand campaign did not exist. Meanwhile, non-brand campaigns targeting high-intent commercial queries were underfunded relative to their pipeline contribution because their CPA looked worse than brand on a last-click basis. B2B marketing benchmarks reveal where the actionable signal lives — and without them, budget follows historical inertia rather than current performance.

Finding 10: There Was No Connection Between Google Ads and the CRM

In fewer than 20% of the accounts we audited was there a working connection between Google Ads click data and CRM deal data. The rest were operating with a fundamental visibility gap: they knew which keywords generated clicks and which generated form fills, but they had no idea which keywords generated pipeline and which generated revenue.

This gap makes it impossible to answer the question that actually matters in B2B paid search: which keywords are producing customers, not just leads? In accounts where we helped establish this connection — even imperfectly, through UTM tagging and manual CRM reporting — the budget reallocation decisions that followed consistently produced 20–40% improvement in cost per pipeline opportunity within one quarter. Not from new campaigns or new creative. From stopping spend on keywords that were producing leads that never closed and concentrating it on the ones that were. A scalable lead generation system requires measurement that connects spend to closed revenue — and in Google Ads specifically, the absence of this connection is the single most expensive gap in the average B2B account.

What the Best Accounts Had in Common

Across the accounts that were performing well — producing qualified pipeline at sustainable CAC with clear measurement connecting spend to revenue — the differentiators were consistent and had nothing to do with budget size or industry.

They had clean conversion tracking with offline data imported. They reviewed search terms weekly and maintained aggressive negative keyword lists. Their landing pages scored above 70 on mobile and matched ad copy precisely. They had separated brand from non-brand campaigns and knew the incremental value of each. They were optimizing toward revenue events, not form fills. And they had a direct connection between Google Ads and their CRM that let them see, at the keyword level, which spend was producing customers.

None of this is technically complex. All of it requires discipline — consistent maintenance, a measurement infrastructure built before it is needed, and a willingness to make allocation decisions based on data rather than dashboard appearances. Growth metrics that predict sustainable growth are the ones that connect activity to revenue — and in Google Ads, the accounts that build that connection consistently outperform those that do not, regardless of budget.

The Bottom Line

The problems we found across 20+ B2B Google Ads accounts were not exotic. They were not the result of negligence or incompetence. They were the predictable result of accounts that had been set up, launched, and then managed reactively — responding to obvious problems while missing the structural issues quietly consuming budget below the surface. If your Google Ads are producing leads your sales team cannot close, your CPA is climbing without a clear explanation, or your account has not had a systematic audit in the past six months, the findings above are a reasonable starting point for understanding why.

Frequently asked questions

How do we know which conversion event to optimize toward in Google Ads if our sales cycle is 3-6 months long?
+
For B2B sales cycles longer than 90 days, optimizing Smart Bidding toward a bottom-funnel event like Closed Won is mathematically impractical — Google needs 30-50 conversions per month per campaign to exit the learning phase. The practical solution is to build a conversion value ladder: assign weighted micro-conversions (e.g., demo request = 10, MQL = 30, SQL = 100) and import offline conversion data from your CRM via Google’s offline conversion import or a tool like Zapier or Salesforce integration. According to Google’s own guidance, accounts using imported offline conversions tied to pipeline stage see 20-30% improvement in lead quality within 60-90 days of implementation. The key constraint is volume — if a single campaign generates fewer than 15 SQLs per month, you likely need to consolidate campaigns before this approach works. Start with demo requests as your primary event only if you’ve verified that at least 25% of those demo requests are converting to pipeline within your CRM.

What is a realistic Quality Score benchmark for B2B Google Ads, and how much does a low Quality Score actually cost us?
+
In B2B paid search, a Quality Score of 7 or above on exact and phrase match keywords is a realistic and achievable benchmark for well-managed accounts — scores of 4-5 on core commercial keywords are a signal of structural problems, not just copy issues. Google’s auction model applies a multiplier to your bid based on Quality Score: a keyword with a QS of 5 can cost you up to 40% more per click than a competitor bidding the same amount with a QS of 8, according to documented Ad Rank calculations. Across the accounts we typically see audited, the average cost inflation from suppressed Quality Scores on high-intent keywords runs $8,000-$25,000 in wasted spend annually at the $15,000-$50,000 monthly budget range. The primary driver of low Quality Scores in B2B is almost always landing page relevance, not ad copy — teams send traffic to generic product pages instead of intent-matched landing pages, which tanks expected CTR and post-click signals. Fix landing page alignment before investing in ad copy iteration.

How do competitor and branded keyword strategies typically fail in mid-market B2B accounts, and what should we actually be doing?
+
The most common failure is running branded campaigns without a negative keyword strategy that excludes navigational queries from non-ICP segments — employees, competitors, and existing customers clicking branded terms inflate conversion counts without contributing pipeline, which then corrupts Smart Bidding signals. A correctly structured branded campaign should be running as a separate campaign with Exact and Phrase match only, with a target CPA or ROAS bid strategy calibrated to your average deal size, not blended with non-brand traffic. On competitor bidding, Wordstream data suggests competitor keywords in B2B typically convert at 30-50% lower rates than branded or category keywords, but are useful for specific account-based plays rather than broad awareness. If you are spending more than 15% of your paid search budget on competitor terms without an ABM list filtering who sees those ads, you are likely generating noise, not pipeline. The highest-ROI branded play most mid-market accounts underutilize is bidding on your own brand name to control the SERP narrative and capture bottom-funnel users — especially critical if you have aggressive review sites like G2 or Capterra bidding on your brand.

Our cost per lead from Google Ads looks reasonable, but sales says the leads are bad. How do we diagnose whether this is a targeting problem or a handoff problem?
+
Start by pulling a 90-day lead sample from Google Ads and running it against your ICP definition across three dimensions: company size, industry vertical, and job title or seniority level — if fewer than 40% of inbound leads match your ICP on all three, you have a targeting problem, not a handoff problem. According to a 2023 Forrester study, 57% of B2B marketers report that lead quality, not volume, is their primary challenge, and the root cause in paid search is almost always keyword intent mismatch combined with broad audience targeting. Specifically, check whether Performance Max campaigns are running alongside your Search campaigns — PMax frequently cannibalizes branded and high-intent traffic while routing budget to Display and YouTube inventory that generates form fills from out-of-ICP audiences. If your ICP match rate is above 40% but sales is still rejecting leads, the issue shifts to SLA and follow-up speed: Drift and Salesloft data consistently shows that B2B leads contacted within 5 minutes are 9x more likely to convert to a qualified conversation than leads followed up after 30 minutes. Run both diagnostics in parallel before blaming the channel.

Should mid-market B2B companies be using Performance Max campaigns, and what are the actual risks of running them?
+
For most mid-market B2B companies spending under $100,000 per month, Performance Max should be approached with significant caution — the campaign type is optimized for conversion volume, which in B2B means it will find the path of least resistance to a form fill, not to an SQL. The structural risk is opacity: PMax does not expose keyword-level search term data, audience segment performance, or placement-level reports in the same way Search campaigns do, which makes it nearly impossible to audit what is actually driving spend. Google’s own internal data shows PMax performs best when an account has 50+ conversions per month feeding the algorithm — below that threshold, the model lacks sufficient signal and defaults to broad reach, which in B2B often means consumer or SMB traffic. If you choose to run PMax, the minimum control architecture should include a comprehensive negative keyword list applied at the account level, an audience signal built from your CRM customer list, and a separate Search campaign running simultaneously so you can compare lead quality between the two. Accounts that replaced all Search campaigns with PMax in 2022-2023 and then audited results 6 months later consistently found CPL decreased while SQL-to-lead ratio dropped by 35-50%, which is a net loss at mid-market deal sizes.

Brent Nakagawa
About the author

Founder & Principal Consultant, Gawa Growth

Brent Nakagawa is the founder of Gawa Growth, a growth marketing consultancy running strategies across paid media (Google, Meta, LinkedIn, Bing, programmatic), SEO, GEO, ABM, demand gen, content, and CRO — for B2B, B2C, local services, and e-commerce businesses.

Growth Marketing Paid Media SEO & GEO ABM Attribution CRO Demand Gen