SIGNAL · Issue 02 — Paid Search · Google vs Bing · Channel Mix
SIGNAL is Ziggy’s field-notes series on the experiments we run and what the data tells us. This issue is anonymised; pipeline is shown only as ratios or percentage change, and all figures are reported exactly as measured.
There’s an unwritten hierarchy in B2B paid search. Google is where the serious budget goes; Bing — Microsoft Ads — is the afterthought you switch on “for incremental reach” and never really look at again. Budgets reflect the bias: across our book, Google takes roughly seven of every eight paid-search dollars. So we did the thing the hierarchy discourages — we normalised our portfolio to the accounts running both platforms, isolated genuine paid-search conversions, and compared the two channels properly. The result doesn’t fit the hierarchy.
This isn’t a story about cheaper clicks. It’s about what those clicks turn into — opportunities, pipeline and won deals — and the fact that, for most of the accounts we tested, the cheaper, less-loved channel is now producing them more efficiently.
−77% Bing cost per opportunity vs Google | ≈ 2× Bing pipeline ROAS vs Google (10.1x vs 5.1x) | −60% Bing cost to acquire a customer vs Google |
The aggregate: cheaper opportunities, stronger pipeline return
Pooled across the accounts running both platforms over the last eight months, Bing produced an opportunity for roughly a quarter of Google’s cost, and returned about twice the pipeline for every dollar spent. The cost-per-opportunity gap is the headline: it’s the metric least dependent on any single account, because the opportunity volume is spread across the book.

Clients running both platforms, paid search, Nov 2025–Jun 2026. Bing’s cost per opportunity is 77% below Google’s; its pipeline ROAS is roughly double. Pipeline expressed only as a return ratio — absolute values withheld.
The cheaper click was the least interesting part. What mattered was that the cheaper click turned into a cheaper opportunity.
The numbers — Bing vs Google (paid search, Nov 2025 – Jun 2026)
| Metric | Bing | Bing advantage | |
|---|---|---|---|
| Opportunities | 494 | 296 | — |
| Cost / Opportunity | £2,347 | £544 | −77% |
| Pipeline ROAS | 5.1x | 10.1x | ≈ 2× |
| Cost to acquire (CAC) | ~£21,500 | ~£8,500 | −60% |
Aggregate across the accounts running both platforms. Pipeline shown as ROAS ratio only; absolute pipeline withheld.
The honest reading: Google still owns the volume — this is an efficiency shift, not a takeover
Two things are true at once. Bing is now the more efficient paid-search channel for most of the accounts running both. And Google is still, by a distance, the bigger one: across these accounts it carried roughly seven times the spend and the large majority of opportunities and won deals. You don’t pause Google on the back of this. And the magnitude deserves caution — the blended ROAS leans toward the largest account’s pipeline, it isn’t universal, and a couple of accounts carry attribution gaps. The robust claim is the cost-per-opportunity one: Bing produces a qualified opportunity for a fraction of Google’s cost.
Why this happens: less competition, a different audience — and a Google SERP in flux
Our working explanation has two parts. First, Bing’s paid-search auctions are far less contested in B2B, so clicks are cheaper — but the bigger effect is downstream. Bing’s search audience skews older, more desktop-led and more enterprise, and for the right offer that audience converts to qualified opportunities at a better rate. So the advantage compounds from a modestly cheaper click into a much cheaper opportunity.
Second — and this is the part most marketers are underweighting — Google’s results page itself has become a moving target. Over the past year Google has been folding AI Mode into an already ever-changing SERP, and that has materially destabilised the volume — and the predictability — flowing through Google. As answers get summarised at the top and the layout keeps shifting, paid listings compete for a smaller, more volatile pool of clicks, which pressures both cost and consistency. Bing, far less disrupted, has been a comparatively stable place to buy an opportunity. Where Bing doesn’t work for an account, it’s usually because almost none of that account’s real pipeline is attributable to paid search at all — most of its demand is direct and organic.
The transferable lessonIf you run Google-only because “that’s where the volume is,” you may be leaving your most efficient pipeline untested — especially as Google’s SERP keeps shifting under you. Test the challenger on every account, judge it on cost-per-opportunity and pipeline ROAS, not CPC or clicks.
What we’re watching next
- Durability at scale. Bing’s ratios are partly flattered by small spend. The real test is whether the efficiency holds as we shift more budget into it.
- Closed-won maturity. Pipeline is a leading indicator; we’re tracking won-deal rates and CAC as these cohorts mature.
- The AI-Mode effect. We’re monitoring how Google’s evolving SERP reshapes paid-search volume and cost from here — and how much demand it pushes toward other channels.
Methodology. Figures are drawn from Ziggy’s live cross-client data warehouse (BigQuery), covering the accounts running both Google and Microsoft (Bing) Ads, Nov 2025–Jun 2026. Paid-search conversions are isolated on first-touch medium = “paid search” and split to channel by source; spend and clicks come from the ad-platform records. ROAS = pipeline ÷ spend; Cost/Opp = spend ÷ opportunities; CAC = spend ÷ closed-won. Aggregate ROAS is pipeline-weighted (so it leans toward the largest account); cost-per-opportunity is opportunity-weighted. Absolute pipeline values are withheld and shown only as ratios. One account’s Google spend is not recorded; pipeline reflects open/expected pipeline; identities are anonymised and currencies assumed comparable.
We re-test the obvious so you don’t have to assume it.
Ziggy is the B2B demand generation agency behind some of the world’s largest technology companies. If your paid search budget split hasn’t been challenged with funnel-level data lately, that’s a conversation we like having. Schedule a call →