B2B Ad Campaigns: Real Test Results from 7 Channels with Pranav Piyush

#259 Paid Ads | In this episode, Dave is joined by Pranav Piyush, founder and CEO of Paramark, a platform helping B2B marketers run real experiments to measure ad performance. Pranav brings a sharp point of view on attribution, channel performance, and how to actually prove what’s working across your paid media mix.Dave and Pranav cover:Real-world results from 7 B2B ad campaigns, including branded search, YouTube, billboards, and Performance MaxWhy most marketers are misusing the word “test” and how to run true experiments with lift, control, and causalityHow even small teams can apply experiment design (on a $10K budget or less) to make smarter spend decisionsWhether you're managing a big budget or just getting started with paid campaigns, this episode will help you think more critically, and confidently, about where and how to invest in B2B marketing.Timestamps(00:00) - – Intro (02:00) - – Why Paramark pulled real data from 7 B2B ad campaigns (04:30) - – Attribution vs. experimentation: what most marketers get wrong (08:00) - – Correlation vs. causation explained (with a LinkedIn example) (10:45) - – How to run a real test (hint: you need a control) (12:00) - – Branded search test results: $1M+ saved, no performance loss (17:00) - – Why strong SEO makes or breaks branded search tests (18:00) - – Billboard test: how one brand proved real lift with out-of-home (21:30) - – What “digital out-of-home” looks like in B2B (23:00) - – YouTube ad tests: one big win, one big flop (26:00) - – How to run tests with small budgets ($500–$10K) (30:30) - – Connected TV (CTV) test results from a Series F SaaS brand (32:30) - – What happens when a multichannel test works—but isn’t efficient (34:30) - – Paramark’s Exit Five sponsorship test (real numbers shared) (38:00) - – Why content needs to drive short-term lift, not just long-term “brand” (41:00) - – How Pranav used LinkedIn to drive inbound from day one (43:00) - – Your attribution model is lying, give your audience more credit (44:30) - – When 7 ad tests fail in a row…and the 8th one works (46:00) - – Performance Max test: why it worked for one brand (48:00) - – How long to run a test? Use data, not gut (50:00) - – Bonus: Pranav’s hiring playbook for his first marketing leader (54:00) - – Wrap up and final takeaways Join 50,0000 people who get Dave's Newsletter here: https://www.exitfive.com/newsletterLearn more about Exit Five's private marketing community: https://www.exitfive.com/***Brought to you by:Optimizely - the AI platform for marketers. Build your own AI agents or pull from a directory of 50+ pre-built ones for marketing use cases. Their new Virtual Teammates can join meetings, complete tasks, support campaigns, and keep your website optimized. Learn more at optimizely.com/exitfive.Webflow - A website platform built for the agentic web, letting modern marketing teams build fully custom sites that perform in AI search with no developer needed. Learn more at webflow.com/for/exitfive.Zoom Webinars & Events – The virtual event platform built to help B2B marketers run webinars that actually drive pipeline, with branded registration pages, live engagement features, and built-in tools to repurpose sessions into clips and content. Learn more at zoom.com/exitfive.Compound Growth Marketing - A full-funnel demand gen agency helping high-growth cybersecurity and enterprise software companies show up earlier in the buying journey, combining AEO, modern paid advertising, and a dedicated go-to-market engineering team. Podcast listeners get two free media planning sessions to find out what channels are driving the best ROI. Learn more at compoundgrowthmarketing.com/exitfive. ***Thanks to my friends at hatch.fm for producing this episode and handling all of the Exit Five podcast production.They give you unlimited podcast editing and strategy for your B2B podcast.Get unlimited podcast editing and on-demand strategy for one low monthly cost. Just upload your episode, and they take care of the rest.Visit hatch.fm to learn more