---
title: "B2B Google Ads Strategy | Three-Pillar Framework | VertoDigital"
description: "Most B2B Google Ads accounts are optimised for leads, not revenue. Here's the three-pillar framework we use in every account audit, and why we only run Search."
url: "https://vertodigital.com/content-hub/blog/b2b-google-ads-strategy"
image: "https://vertodigital.com/images/site/og-default.png"
---

# B2B Google Ads strategy: a three-pillar framework for campaigns that actually drive pipeline

Most B2B Google Ads accounts are optimised for leads, not revenue. Here is the three-pillar framework we use in every account audit, and why we only run Search.

*By Ivailo Shipochki, Partner, Inbound & Outbound Growth, VertoDigital · Last updated Aug 13, 2026 · 13 min read*

## Key takeaways

- Search only. We run Google Ads as a Search-only channel. Display, Video, Performance Max, and Smart campaigns get disregarded in B2B accounts because none of them give a buying committee the targeting control it requires.
- TOF belongs on LinkedIn. Top-of-funnel and awareness budget performs better there, with full control over firmographic and job-title targeting plus impression-level pipeline attribution.
- Diagnose first. Every audit starts with one number: Spend Under Risk, the share of budget sitting in non-search inventory, Dynamic Search Ads, or broad match keywords with no real targeting control.
- Sequence matters. Real performance comes from three pillars, Structure, Signals, Optimisation, done in that order. Skipping ahead to smarter bidding before fixing structure and signals is why most B2B accounts plateau.
- Report on pipeline, not CPL. Cost per lead is not a commercial metric. Accounts that move revenue are managed against CRM lifecycle stages, MQL, SQL, Opportunity, Closed-Won, not form fills.

**Best for:** B2B SaaS and technology marketing leaders, demand generation leaders, paid media managers, and revenue teams responsible for turning Google Ads spend into qualified pipeline.

A B2B Google Ads strategy drives pipeline when it prioritises high-intent Search demand, uses CRM-qualified conversion signals for bidding, and measures performance through MQL, SQL, opportunity, and closed-won revenue rather than cost per lead alone.

Most B2B Google Ads accounts look healthy on the platform and hollow in the CRM. Impressions climb, CTR looks fine, cost per lead sits inside target. Then you open the CRM and the story falls apart: few of those leads ever turn into pipeline.

That gap between what the platform reports and what the business actually closes is the first thing we look for in every account we audit. It's rarely a bidding problem. It's almost always a structure problem, a signal problem, or both, and fixing it starts with a decision most agencies won't make.

## Why Search is the core B2B Google Ads channel

Search captures real-time, ICP-level transactional intent. Someone is actively typing what they need, right when they need it. It's the strongest direct hand-raiser channel in a B2B media plan, and usually the paid channel most directly tied to pipeline contribution, because the query itself carries proof of intent that no audience-targeting parameter can replicate.

Everything else Google Ads sells, Display, Video, Performance Max, Discovery, Smart campaigns, Dynamic Search Ads, asks you to trade that intent signal for reach. In consumer categories, that trade can work. In B2B, it rarely does.

None of that inventory gives you real control over who the ad actually reaches. Performance Max in particular is built to obscure placement and audience data in exchange for automated reach. That's the opposite of what a committee-based, long-cycle B2B sale needs.

So the accounts we run are Search-only. When a client wants top-of-funnel reach among a specific set of job titles, industries, and company sizes, that budget goes to [LinkedIn](/content-hub/blog/linkedin-ads-vs-google-ads), not Google's Display Network or Performance Max.

LinkedIn is the platform where we retain full control over who sees an impression and can tie that impression back to pipeline influence. Google Ads' job is to capture demand that already exists. LinkedIn's job is to help create it. Blur that line and B2B budgets end up funding reach that never converts.

### How we identify Google Ads spend that is under risk

Once we know where the risk sits, the first fix is structural.

## Pillar 1: campaign structure

Two guardrails, borrowed from Google's own smart bidding guidance and not something we invented: ad groups tend to need somewhere in the neighbourhood of 3,000 impressions a week, and campaigns need something like 30 conversions a month, before an automated bidding algorithm has enough signal to optimise reliably. Treat these as a design target, not a hard law. The point isn't hitting an exact number. It's recognising that an ad group split too finely to ever reach meaningful volume can't generate a usable signal, no matter how it's bid.

Most B2B accounts we inherit are split the opposite way: dozens of narrow ad groups, each built around a handful of near-duplicate keywords, each starved for volume. Consolidating those into fewer, broader ad groups is the fix. But aggregation done carelessly trades relevance for volume, which just relocates the underperformance instead of solving it. Two things keep relevance intact at scale:

- **Dynamic ad attributes:** these let ad copy adjust to the specific query or audience within a consolidated ad group, instead of running one generic message across everything.
- **Keyword-level landing pages:** these preserve message match at the page level even after the ad group itself has been broadened.

We also check impression share before deciding whether an ad group is a consolidation candidate or a genuine low-demand category.

| What you see | What it usually means | Recommended action |
|---|---|---|
| High impression share, above roughly 60%, and low weekly volume | The ad group has reached the ceiling of its available market. | Expand into relevant adjacent keywords. Do not simply add more budget. |
| Low impression share, below roughly 30%, and low weekly volume | Available demand is not being captured. | Review budget, bids, ad rank, and eligibility constraints. |
| Low impression share in branded ad groups | Brand demand may be constrained. | Review budget allocation, negative keywords, and campaign settings. |

Those should show high impression share by default, and if they don't, that's usually a budget or negative-keyword issue worth fixing right away.

## Pillar 2: conversion signals

Structure only matters if the algorithm is optimising toward something that reflects real commercial value.

Most B2B accounts are still bid against a single conversion event: the form fill. That is a problem. A form fill and a qualified opportunity are not the same event, and an algorithm trained on the former will happily buy you more of it because it has no way of knowing the difference.

That is consistent with what we see across the accounts we onboard: roughly 70% are not leveraging offline conversions in any form.

The goal is to give Google Ads signals that reflect commercial quality, not simply lead volume. This starts with two conversion types, followed by the technical setup needed to get those signals back into the platform.

### Step 1: add two conversion signals that reflect lead quality

| Conversion type | What it is |
|---|---|
| **CRM lifecycle conversions** | Import MQL, SQL, Opportunity, and Closed-Won events from the [CRM back into Google Ads as offline conversions](/content-hub/blog/crm-offline-conversion-tracking-b2b-advertising), each weighted appropriately. This is the single highest-impact change available in most accounts we audit, and also the one most have not made. The majority of accounts we look at have no CRM or offline conversion import configured at all. |
| **Custom ICP Lead conversion** | Create a filtered conversion definition built on firmographic signals captured at the point of form submission, such as job-title match, company size, and similar criteria. This acts as a quality gate before CRM data has had time to mature. It gives the algorithm a directionally better signal from day one, instead of waiting weeks for a lead to move through the CRM pipeline. |

### Step 2: capture and persist the Google Click Identifier

Setting either conversion signal up starts with the Google Click Identifier (GCLID).

GCLID gets captured the same way a lead form already captures UTM parameters: as a hidden field. The difference is that GCLID lives in the page's query string. If a visitor moves to a second page before filling out the form, the parameter is gone unless you persist it across the session first.

The setup sequence is:

1. Capture GCLID as a hidden field when the visitor arrives from a Google Ads click.
2. Persist it across the session.
3. Store it against the lead record in the CRM.
4. Use it to match later CRM lifecycle events back to the original Google Ads click.

Without this, later-stage CRM outcomes cannot be reliably attributed back to the original Google Ads interaction.

### Step 3: choose how CRM conversion data returns to Google Ads

From there, choose how the data moves from the CRM back to Google Ads.

HubSpot, Marketo, and Pardot all offer native connectors, and they are the easier setup. But they only support lifecycle-stage triggers, give you no visibility into import error logs, and, because of how each platform's conversion API handles timestamps, typically land a match rate somewhere between 50% and 80%.

As explained in our [Offline Conversions for B2B Inbound ebook](/content-hub/ebooks/offline-conversions-b2b-inbound), this match rate can mean that a meaningful share, and in some cases as much as half, of the CRM conversions you expect to feed into Google Ads never reach the platform or influence bidding.

Native connectors can be useful for lifecycle-stage triggers, but they should not be treated as set-and-forget. Monitor the number of eligible CRM conversions successfully matched back to Google Ads clicks. Investigate unmatched records. Confirm that conversion timestamps and click identifiers are being passed correctly. A fully monitored offline-conversion workflow gives more visibility and control over the data returning to Google Ads.

### Step 4: use lifecycle-stage conversions only when they are bid-ready

A lifecycle-stage conversion is only worth using if it clears three bars:

- Fewer than 15 days between click and conversion
- More than 50 monthly attributed conversions
- Lead scoring that already accounts for ICP fit, so unqualified leads do not get counted

Most MQL definitions fail at least one of these. That is exactly why the custom ICP Lead conversion exists. It is built on firmographic attributes instead of a behaviour score, so it can fire the same day a lead lands in the CRM. This gives Google Ads an earlier quality signal while lifecycle-stage data matures.

### Step 5: check which conversions are actually influencing bidding

Once both conversion types are in place, flag any conversion action with a wide gap between total conversions and conversions actually included in bidding. That gap usually means something is being tracked but never fed to the algorithm. A conversion that appears in reporting but is excluded from bidding cannot improve the quality of traffic Google Ads acquires.

### What this looks like in practice

We have seen this play out at scale. At [IRONSCALES](/case-studies/ironscales-integrated-reporting), an enterprise email security company, we paired this same CRM-based signal set with unified reporting across every paid channel. Paid-channel pipeline volume grew 440%, a 4.4x increase, over the following year.

The signal chain is:

Google Ads click -> GCLID capture -> form submission -> CRM qualification -> offline conversion import -> bidding signal -> pipeline measurement

## Pillar 3: smart bidding

Smart bidding is the last pillar, not the first. Running it out of order is the most common mistake we see. An automated bid strategy trained on a fragmented account structure and form-fill conversions will optimise beautifully toward exactly the wrong outcome. The sequence matters:

1. Migrate to the aggregated, signal-sufficient structure (Pillar 1).
2. Import CRM lifecycle conversions and activate the custom ICP lead conversion (Pillar 2).
3. Switch bidding to Maximize Conversions or Target CPA against the new signal set, not the old form-fill-only conversion.
4. Once that's stable, layer in conversion value weighting and move toward Maximize Conversion Value or Target ROAS, so the algorithm bids harder for opportunities and closed-won revenue, not just any conversion.

Where the client's stack supports it, the same CRM signal set extends to Microsoft Advertising as a secondary search channel. The logic doesn't change. Only the platform does.

## Measure the account against pipeline, not cost per lead

None of the above matters if the reporting layer still stops at cost per lead. Accounts that hold leadership's confidence report on the same chain the CRM already tracks, with cost and CAC attached at every stage: Search query -> Lead -> MQL -> SQL -> Opportunity -> Closed-Won.

A campaign with a high cost per lead and a strong SQL rate is healthier than one with a cheap cost per lead and nothing behind it. But you can only see that if branded and non-branded performance are reported separately, since brand efficiency, often artificially cheap and high-converting, otherwise masks non-brand underperformance in the blended number.

## A realistic timeline

We sequence the three pillars across roughly a 90-day window instead of compressing them:

- **Days 1-30 (Foundation):** run the audit, quantify Spend Under Risk, consolidate campaign structure, and remove non-search inventory, DSA, and low-discipline broad match.
- **Days 31-60 (Signals):** connect the CRM, import lifecycle conversions, and launch the custom ICP lead conversion.
- **Days 61-90 (Optimisation):** shift bidding to the new signal set, validate performance against SQL and Opportunity data, and expand.

An account that skips straight to Pillar 3 without doing 1 and 2 usually looks like it's improving for the first few weeks. Then it plateaus once the algorithm exhausts the shallow signal it was given.

**What should I do now?** Request a [Pipeline Assessment](/assessment) to identify where Google Ads spend, campaign structure, or conversion signals may be limiting qualified pipeline. Explore our [case studies](/case-studies) to see how B2B and SaaS teams have improved pipeline visibility, measurement, and paid media performance. Or explore [B2B and SaaS Paid Search](/services/inbound/b2b-and-saas-paid-search) to see how we build Search programmes around qualified pipeline rather than form fills.

## FAQ

### Should B2B companies run Performance Max or Display campaigns on Google Ads?

Generally no. Both trade transparency and targeting control for reach, and B2B buying committees are narrow enough that the loss of control usually costs more in wasted spend than the reach is worth. If reach and awareness are the goal, that budget is typically better spent on LinkedIn.

### What's a realistic Google Ads budget for a B2B or SaaS company?

It depends heavily on category and CPCs, but as a rough floor, most B2B search programmes need something in the neighbourhood of $20,000+ in monthly spend before there's enough volume across campaigns to run a real signal-based optimisation strategy. Below that, the priority is keyword and structure discipline over automation.

### Where should top-of-funnel and awareness budget go instead of Google Ads?

LinkedIn. It's the platform where B2B advertisers retain full control over firmographic and job-title targeting, and where impression-level exposure can be tied back to pipeline influence, something Google's non-search inventory doesn't offer at the same level of precision.

### How much weekly impression volume does an ad group need before smart bidding works?

Roughly 3,000 impressions a week is the benchmark Google's own smart bidding team cites as a reasonable floor for reliable signal. Treat it as a design target, not a strict cutoff. An ad group split too finely to reach meaningful volume can't generate a usable signal, regardless of the bid strategy applied to it.

### Do we need CRM integration before Google Ads can drive pipeline instead of just leads?

Not on day one, but it should be one of the first structural changes made. Until [CRM lifecycle data](/content-hub/blog/rebuilding-attribution-closed-won-not-mqls), MQL, SQL, Opportunity, Closed-Won, is flowing back into Google Ads, the algorithm can only optimise toward form fills, a weak proxy for commercial value.

### How long before a restructured B2B Google Ads account shows pipeline impact?

Most of our engagements run structure, signal, and optimisation changes across a 90-day window, with the clearest pipeline impact showing up once the bidding strategy has had close to a full sales cycle of CRM-informed data to learn from. Timing varies by deal cycle length.