← Back to blog

Google Ads for B2B — Why It Needs a Different Approach Than B2C

Five things that genuinely make B2B different from B2C

This is not "the same campaign with different keywords". It is a different model. Understanding the differences is the foundation — without it, the rest of this article stays theoretical.

1. Decision cycle: 3–9 months instead of minutes

A B2C customer buys impulsively or after short deliberation. B2B goes through research (compares vendors), evaluation (testing, demo), proposal (pricing, negotiation), procurement (approvals, legal). It takes months. For your campaign that means a click from Monday becomes a customer in September. The standard 30-day attribution window in Google is not enough — it has to be extended to 60–90 days, and externally (in CRM) to 180+ days.

2. Multiple decision-makers (Buying Committee)

An enterprise SaaS purchase often involves 5–7 people: end user, IT, security, finance, procurement, C-level sponsor. Each one clicks from a different device, in a different place, on a different keyword. That means one lead in CRM = dozens of clicks in Google Ads. Attribution becomes complex.

3. Lead ≠ customer. Lead ≠ MQL. MQL ≠ SQL. SQL ≠ deal.

In B2C, a conversion = revenue. In B2B, a conversion on the website (form, demo request) is just the first step in the pipeline. From my observations, typical conversion rates in Polish B2B SaaS:

• Visit → MQL (Marketing Qualified Lead — filled in a form): 1–4%
• MQL → SQL (Sales Qualified Lead — sales wants to talk): 15–35%
• SQL → Opportunity (there is a budget, project, decision-maker): 25–45%
• Opportunity → Closed-Won: 15–35%

From 1,000 visits you get 30 MQLs → 8 SQLs → 2.5 Opportunities → 0.7 deals. These are typical rates, not anomalies. Every fix at any stage is leverage on the entire funnel.

4. Contract value breaks CPL logic

CPL of 800 PLN sounds expensive. But if a B2B customer has an LTV of 80,000 PLN (e.g. a 3-year SaaS subscription at 2,200 PLN per month), then with an MQL → customer rate of 8% the CAC works out to 10,000 PLN — less than 13% of LTV. Very healthy. What's more, in B2C you optimize for CPA, in B2B increasingly for cost per opportunity or cost per Closed-Won with value-based bidding (tROAS on deal value).

5. Low search volume on niche keywords

"divorce lawyer Krakow" — several thousand searches per month. "enterprise data lake migration consultancy" — a few dozen. In niche B2B, Google queries have an extremely low volume — sometimes 5–20 searches per month on a specific combination. That means Search alone is not enough — you need LinkedIn, content, account-based marketing.

Polish benchmarks — Hotlead — B2B lead costs by channel 2026, Artur Smolicki — Google Ads for B2B. Global — Growthspree — SaaS Google Ads Benchmarks 2026.

Step 1 — what you need BEFORE you turn on Google Ads in B2B

This is the list that is most often missing from failing B2B accounts.

Value-based conversions — not just "form fill"

The worst B2B setup I see regularly: a single "Lead" conversion with value 1, hit by everyone — a student downloading a whitepaper and an enterprise CTO clicking "demo request". The system does not differentiate. It optimizes for volume, finds the students. The money flows out.

The setup that works — different conversion types with different values. Note: the values below are an arbitrary point scale, not PLN — what matters are the proportions between them (that an SQL is 5× more important than an MQL), not the specific numbers. The exception is Closed-Won, where it is worth using the actual deal value.

• Newsletter / whitepaper — value 5–10 (means "someone is aware of the problem" but far from buying)
• Demo request / sales contact — value 50–100
• MQL (qualified by marketing) — value 100–300
• SQL (qualified by sales) — value 500–1,500
• Opportunity (deal in pipeline) — value 2,000–5,000
• Closed-Won — actual deal value (e.g. 50,000 PLN)

Smart Bidding understands the hierarchy and starts looking for people who convert deeper into the funnel, not just on the top.

Offline Conversion Import (OCI) — a game changer in B2B

Most B2B conversions do not happen on your website. The customer clicked the ad 3 months ago, filled a demo, sales sent an offer, someone signed a contract — all of that outside Google Ads. Without OCI, Google does not know that this click paid off. It only sees the "Demo request" form from 3 months ago.

How it works technically:

1. Customer clicks the ad → Google saves the GCLID (a unique click identifier).
2. Your system (form, CRM) captures the GCLID from the URL and stores it on the new lead.
3. When the lead advances in the pipeline (MQL → SQL → Closed-Won), the CRM sends an update to Google Ads with the GCLID + date + new value.
4. Google Ads gets the signal: "this June click was a 80,000 PLN deal" — and starts looking for similar clicks.

HubSpot has built-in ad conversion event sync for this (check the Ads → Events section in your portal — the UI naming changes between versions). Salesforce — via Google's Ads Data Manager connector (Google Ads → Tools → Ads Data Manager). Without a CRM, you only need a form on the website + a script saving the GCLID + a cron job exporting changes to the Google Ads API. From my experience and Revvim's data — campaigns that implemented OCI with SQL and Closed-Won signals see a clearly better lead quality (typically 20–50% improvement in MQL → SQL rate at similar volume). That is not always +50% SQL in absolute numbers, more often a drop in cost per SQL at the same volume — but the net effect is the same.

OCI mechanism and typical uplift — Revvim — Bidding on Offline CRM Stages, Google Ads OCI Setup Guide 2026.

Attribution window 60–90 days (not the default 30)

In Google Ads account settings there is a conversion window — by default 30 days. For B2B that is too short. Set 60 days for MQL, 90 days for SQL/Opportunity. The GCLID is active for up to 90 days for standard OCI imports — for enterprise cycles > 3 months, you have a gap. Workaround: Enhanced Conversions for Leads based on a hashed email or customer identifier — they work beyond the GCLID limit. That is a separate technical topic, but it's worth knowing the gap is solvable.

Conversion volume — do you even have data to optimize on

Smart Bidding needs enough conversion signal to learn — roughly 15–30 value-based conversions per month, more for tROAS. Google no longer publishes a hard threshold (the old numbers came from 2018–2022 documentation), but below that range the algorithm behaves erratically. In niche B2B (5 SQLs per month) there simply is not enough to learn from — it falls back to optimizing for cheap leads. The fix: optimize for conversions earlier in the funnel (MQL instead of SQL), with proportional values. The system has more signals, and you do not lose priorities.

Step 2 — a B2B campaign structure that works

In B2B, less is more. Start with the minimum, add only when the basics work.

Campaign	Why	% of budget	When
Branded Search	Captures decision-makers who already know your brand. Critical — when the competition bids on your name, you lose leads.	10–15%	From day 1
High-intent non-brand Search	Queries like "<solution> pricing", "<product> vs <competitor>", "<industry> software". Highest conversion rate.	50–60%	From day 1
Educational non-brand Search	"how to choose X", "best Y for Z" — top of funnel, low conversion, but nurtures. Essential for building demand.	15–25%	After 30 days (once basics work)
Demand Gen / YouTube	Awareness in niche segments. Expensive, long payback.	0–10%	Only when budget exceeds 15,000 PLN/mo and CRM sees attribution
PMax	In typical B2B services rarely makes sense — no feed, weak signals. Exception: B2B e-com with a feed or advanced Customer Match with deep CRM signals.	0–5%	Test only when Search works and you have at least 50 SQL/mo as a signal

Philosophy: start with high-intent + brand, add educational only when CRM sees pipeline. PMax and Demand Gen in typical B2B services are expensive experiments, not a foundation.

LinkedIn Ads vs Google Ads in B2B — when to use what

A common question from B2B founders. Briefly:

• Google Ads wins when the customer is actively looking for a solution. Examples: "CRM for ad agency", "IT security audit pricing", "outsourced bookkeeping for startup". High intent, low competition in niches.
• LinkedIn Ads wins when you want to reach a specific segment that has not yet defined the problem. Targeting by job title, industry, company size. Good for ABM (Account-Based Marketing) and thought leadership.
• The most common combination I see in successful B2B: Google Search for high-intent queries + LinkedIn for TOFU/MOFU education + cross-platform remarketing.

If your budget is below 5,000 PLN per month, I would usually focus on one platform and do it well — rather than half-doing both.

Step 3 — what to expect and how to read the results

The first 90 days — what to expect

• Month 1: the data is too shallow to draw conclusions. CPL on demo level rises by 30–60% above target because Smart Bidding is learning. Keep your hand off the strategy.
• Months 2–3: the first SQLs come back from CRM. The algorithm starts to understand which clickers convert deeper. CPL stabilizes, lead quality rises.
• Months 4–6: the first Closed-Won from campaigns. You can switch the strategy to value-based bidding (tROAS on deal value). This is where the real optimization begins.

If you judge B2B Google Ads after 30 days — you are judging the learning phase, not the result. That is the most common mistake I see in CFOs looking at monthly reports.

The most common B2B mistakes

1. Optimizing for "cheap leads". Goal: maximize the number of conversions. Effect: the system finds students filling whitepapers. Fix: different conversion values + value-based bidding.
2. No offline conversion tracking. Google does not know which clickers convert deeper. Money goes top of funnel, deals are accidental. Fix: HubSpot/Salesforce integration + GCLID capture.
3. 30-day attribution window. Most B2B conversions happen after 30 days. Google does not see them. Fix: 60–90 days + enhanced conversions.
4. Broad match in niche B2B. The system reads "enterprise CRM" as "CRM" and bids on hobbyists. Fix: phrase / exact match + tight negatives.
5. Short campaign rotation. "After 14 days I see no SQL, I'm turning it off." In B2B that is too soon — the algorithm needs at least 60 days. Fix: patience + budget appropriate to the cycle.
6. No alignment with sales. Marketing delivers leads, sales says they are "weak". Without CRM signal on lead quality (MQL → SQL conversion rate per query) — money flows to queries that drive volume, not deals. Fix: regular sales-marketing meetings + SQL feedback loop.

What I see after 11 years — B2B segments I have run

Reference ranges from various Polish B2B account types:

• B2B mystery shopping (auditor recruitment) — CPC 1.00–1.50 PLN on queries like "mystery shopper job", CPA 5–15 PLN for free candidate registration. Don't compare 1:1 with a sales lead — that's a completely different kind of "conversion".
• B2B mystery shopping (selling a service to companies) — CPC 3.30 PLN, CPL on demo 200–400 PLN, average deal 50–100k PLN annually. CAC ~5–10% of LTV — very healthy model.
• B2B SaaS (cybersecurity / data) — CPC 12–30 PLN on niche queries, CPL 200–600 PLN, decision cycle 3–6 months, deal 30–200k PLN.
• B2B industrial — CPC 4–12 PLN, CPL 150–400 PLN, long cycles (6–12 months), contract value from 50k to several hundred thousand PLN.
• B2B medical services for companies (occupational health) — CPC 3–8 PLN, CPL 80–250 PLN, but the customer buys an annual package for many employees — LTV very strong.

Source: accounts I run at JP Ads (anonymized, industry + scale). Supporting data — Double Digital — Google Ads costs 2026, Factors.ai — Google Ads B2B Strategy 2026.

When Google Ads for B2B does NOT make sense

Four situations where you have to be brave enough to say "don't launch":

• No sales process. The lead lands in a domain inbox and waits. No salesperson, no follow-up, no nurture. Money goes to waste. Fix the pipeline first — then the ads.
• Low contract value (LTV < 5,000 PLN) + narrow demand + long cycle. The math does not work. Better off with content marketing + long-tail SEO + organic LinkedIn.
• No sales content. No persona-specific LP, no case studies, no demo, not even a decent pricing page. B2B buys only after education — without content your site has no way to convert.
• Niche demand + low budget. 5 monthly searches on your queries + a 1,000 PLN budget = no data for Smart Bidding to learn. Better off with account-based marketing through LinkedIn + cold outbound.

How to read the results — what to measure in B2B

In B2C you look at ROAS and CPA. In B2B you have more layers:

Level 1 — is there data

• Conversions like "Demo request" / "Contact" from Google Ads > 15 in 30 days
• OCI is working — you see "Click conversion (Imported)" as a source on conversions
• Each CRM stage (MQL, SQL, Closed-Won) has its own conversion with a value

Level 2 — efficiency at the click level

• CPL on demo / contact within the band for your industry (see pyramid above)
• Quality Score ≥ 6 for the main keywords
• Search lost IS (rank) < 30%
• CTR > 4% on high-intent non-brand queries (branded typically 10–20%; niche B2B with very low search volume may sit at 3–5% — track relative to your own baseline, not an absolute number)

Level 3 — efficiency at the business level

• MQL → SQL conversion rate > 15%
• SQL → Opportunity > 25%
• Opportunity → Closed-Won > 15%
• Cost per Closed-Won < 15% of LTV (healthy CAC/LTV ratio)
• Sales feedback: are leads from Google Ads better / worse / same as from other channels

The most common diagnostic outcome: levels 1 and 2 are OK, but level 3 is failing — there are leads, but their quality is low. That is a signal to change the queries (less educational, more intent-driven), conversion values (a bigger gap between MQL and SQL) or content (the LP is not qualifying on its own).

If you have not yet implemented OCI — don't wait for full automation to start diagnosing. The simplest start: a Google sheet where sales marks every two weeks per query / campaign whether the lead was "good/weak". Even a manual feedback loop signals where the money is going to waste. Automate via OCI later.

About the sources: Polish benchmarks — Hotlead — B2B lead costs by channel 2026, Artur Smolicki — Google Ads for B2B, Double Digital. Global SaaS benchmarks and OCI mechanism — Growthspree, Revvim — Bidding on CRM Stages, Factors.ai. The rest is observations from 11 years of running Google Ads campaigns at JP Ads — numbers from my own practice are flagged "from my experience".