The uncomfortable truth about B2B paid search
Most B2B paid search accounts are optimised for the wrong thing. They're built to generate clicks, impressions, and lead form fills — metrics that look good in a weekly performance report but have no reliable relationship to closed revenue.
If your paid search campaigns are generating 200 leads a month and your sales team closes 2% of them, you're not running a growth engine. You're running an expensive contact database builder.
The problem isn't paid search as a channel. The problem is how most B2B teams apply it.
Why B2B paid search breaks in ways B2C doesn't
B2C paid search works on a relatively simple model: someone searches for a product, clicks your ad, buys the product. The conversion event is clear. The attribution is clean. The optimisation signal is fast.
B2B paid search operates under completely different conditions:
Long buying cycles. Your ideal customer searches today, starts an internal evaluation in six weeks, gets budget approval in Q3, and signs in Q4. Smart Bidding's machine learning has no idea what to do with a 9-month signal lag.
Committee buying. The person who clicks your ad for "marketing attribution software" is probably a marketing ops analyst. The person who approves the purchase is a CMO who never touched a search engine looking for your category. You're optimising for the researcher, not the decision-maker.
Misaligned intent signals. B2B search volumes are tiny compared to B2C. You might have 50 high-intent searches a month for your core category keyword. Google's auction system is trying to learn from a dataset of 50 data points. That's not enough signal to optimise anything.
Lead quality variability. A lead from "enterprise marketing attribution platform" and a lead from "free analytics tool" might both fill out the same form. One is worth $80k ARR. One is worth nothing. Your CPA-optimised campaigns can't tell the difference.
The framework that actually works
Building a B2B paid search programme that generates pipeline — not just leads — requires rethinking three things: what you're bidding on, what you're optimising for, and what success looks like.
1. Intent architecture over keyword volume
Stop thinking about keywords as search terms and start thinking about them as intent signals. Every keyword in your account sits somewhere on a spectrum from "totally off-category curiosity" to "actively evaluating vendors in your space right now."
Your job is to build intent architecture: a structured approach to which keywords you actively invest in, which you monitor without investing, and which you actively exclude.
Tier 1 — Buying intent: These are searches from people actively evaluating vendors or solutions in your category. They contain category-specific terminology, competitor names, or explicit buying signals ("pricing," "demo," "vs," "alternatives"). These keywords deserve maximum investment and aggressive negative keywords to protect the intent signal.
Tier 2 — Problem awareness: These searches indicate the person has identified a problem you solve, but hasn't necessarily framed it in solution terms yet. "How to improve lead attribution" is a Tier 2 search for a marketing attribution platform. These keywords generate volume but require more sophisticated landing page strategies to convert interest into pipeline.
Tier 3 — Informational: High volume, low intent. These searches generate traffic from people who are nowhere near a buying decision. Most B2B paid search accounts over-invest here and wonder why their lead quality is terrible.
2. Optimise for pipeline, not leads
The single most impactful change you can make to a B2B paid search account is connecting it to your CRM and feeding qualified pipeline data back into your bidding strategy.
This requires three things:
Offline conversion tracking. Import your qualified pipeline stages into Google Ads as offline conversion events. When a lead from paid search becomes a sales-qualified opportunity, that event fires in your campaign data. This gives Smart Bidding a real signal to optimise against instead of "form fill."
Lead scoring at the point of form fill. Before you let Google count a conversion, score the lead. A company with 10 employees searching from a free email domain is not the same conversion as a VP of Marketing at a 500-person SaaS company. Build that distinction into your conversion tracking.
Patience with data accumulation. The above only works once you have enough conversion data for Smart Bidding to learn from. In most B2B categories, that means running manual CPC or tCPA with a conservative target for the first 60-90 days while you build your data foundation.
3. Stop measuring what's easy, start measuring what matters
The metrics that dominate most paid search reports — impressions, clicks, CTR, conversion rate, CPA — are all fine as diagnostic tools. They're not strategy metrics. They tell you how your campaigns are performing against their current optimisation target. They don't tell you whether that optimisation target is the right one.
The metrics that matter for B2B paid search:
- Pipeline generated: Total value of sales opportunities created from paid search traffic, by quarter
- Pipeline CPA: Total paid search spend divided by pipeline generated. This is your real cost-per-acquisition
- Pipeline-to-revenue conversion rate: Of the pipeline generated by paid search, what percentage closes? If it's significantly lower than your organic pipeline conversion rate, your lead quality problem is worse than you think
- Category keyword share of voice: What percentage of impressions are you capturing for your most important buying-intent keywords? If a competitor is dominating the intent signals that matter most, that's a structural problem that no amount of bid optimisation will solve
The negative keyword investment no one talks about
If you had to pick one lever that separates high-performing B2B paid search from low-performing B2B paid search, it would be negative keywords.
Most B2B accounts have a few hundred negatives. The accounts generating real pipeline have several thousand. They've done the work of systematically eliminating every search variation that brings in traffic that will never buy from them.
The categories of negatives that matter most in B2B:
- Job seeker traffic: "Jobs," "careers," "salary," "how to become," "certification," "training" — these generate massive irrelevant traffic for almost every B2B category
- Student/academic traffic: "Essay," "thesis," "assignment," "research paper," "example" — same problem
- Free/DIY seekers: "Free," "open source," "DIY," "without" — unless you have a freemium model, this traffic won't convert
- Competitor employees: Your competitors' employees search their own product names constantly. They're not prospects
- Geographic negatives: If you don't sell in certain markets, negative them at the campaign level
What a good B2B paid search programme looks like
By the end of a properly built 90-day programme, you should have:
- A negative keyword library with 2,000+ entries
- Offline conversion tracking connected to your CRM's pipeline stages
- An intent architecture with clearly tiered keyword investments
- Campaign structure that mirrors your buyer's journey, not your product catalogue
- A weekly optimisation cadence that measures lead quality, not just lead volume
- A monthly pipeline review that connects spend back to revenue
---
DemGen Daily covers paid media strategy for B2B marketing practitioners. Explore more PaidLab articles for frameworks, analysis, and tactical guides.
