If your media plan looks like a love letter to the usual duopoly, you're paying a premium for predictability. The fastest way to shave CPAs is to stop treating Meta and Google as the only levers. Reallocate a slice of spend to alternative networks and formats where competition is lighter and attention is cheaper — but don't do it willy-nilly.
Start with a disciplined experiment: earmark a modest exploration budget (think 10–25%) and run 6–8 short tests across native exchanges, programmatic open auctions, CTV pockets, and specialty social platforms. Keep audiences narrow, creatives tailored, and goals binary: did this channel deliver a CPA below your baseline or not? If yes, scale; if not, kill fast.
Measurement wins the day. Use consistent attribution windows, identical conversion events, and simple control groups so you're comparing apples to apples. Layer in automated rules that shift spend away from rising CPAs and toward cost-per-conversion winners by time of day, placement, or creative. Small rules reduce manual bias and stop overspending before it balloons.
Finally, make diversification repeatable: document winning combos, bake them into audience stacks, and rotate fresh creative into cheap pockets every 10–14 days. Over time you'll harvest multiple low-cost sources and reduce dependence on any single platform — which means steadier volume, lower CPAs, and more negotiating leverage when the giants raise their prices.
Most advertisers pour budgets into the same duopoly and wonder why cost per click climbs. Meanwhile, Reddit, Quora, and Pinterest quietly deliver high-intent visitors for a fraction of the CPM. These platforms reward niche relevance: a well placed post in a focused subreddit, a crisp answer on a popular Quora thread, or a scroll-stopping Pin can reach people already primed to act. The trick is matching format to intent, not copying a Facebook playbook.
On Reddit, start by listening. Spend a week reading top posts in target subreddits to learn tone and rules. Then test organic posts that add value before promoting. Use subreddit targeting for promoted posts and craft headlines that feel native rather than clickbait. Budget tiny daily tests, scale winners, and avoid generic creatives that scream ad to mods and users.
Quora is where long tail intent lives. Target specific questions and topics rather than broad audiences, and pair promoted answers with deep, helpful content on your landing page. Bids can be lower than search because you are capturing curiosity rather than interrupting intent. Use answer-first creatives, add a clear next step, and retarget question viewers with tailored offers.
Pinterest is visual search, not social scrolling. Use vertical, lifestyle images, keyword rich descriptions, and Rich Pins or catalog ads if you sell products. The platform surfaces discovery-ready users who are planning purchases, so focus on product intent and clear CTAs. Run small experiments across all three, reallocate from underperforming Meta campaigns, and enjoy the cheap clicks while competitors keep chasing the obvious channels.
Think of Outbrain and Taboola as the discount aisle of paid discovery: lower CPCs, high top-of-funnel intent, and audiences scrolling for answers. Native units don't scream "ad" — they whisper curiosity, which makes them ideal when you want inexpensive clicks that can actually move people down the funnel. Treat them like magazine blurbs, not search snippets: a tight hook, a bold visual, and a single promise that's fulfilled immediately after the click.
Creative is the real conversion lever. Try micro-headline formulas like "You're Doing X Wrong", "How to Get Y in Z", or "Don't Make This Mistake", and pair those lines with a human face or an uncluttered demo image. Optimize tiny thumbnails for contrast and legibility, then match the landing page copy to the ad promise — short, fast, and focused. Sprinkle quick stats or one-line social proof to flip skepticism into curiosity.
Placement and bidding are tactical moves: favor discovery feeds over low-attention widgets, bid CPC to keep acquisition predictable, and segment by interest or referrer. Run creative sprints — test headlines first, then thumbnails, then body — and rotate assets every 3–7 days to avoid creative fatigue. Start with small-budget experiments to find winners, then scale by incrementally raising bids and broadening lookalike-like audiences.
Measure beyond clicks: track micro-conversions (scroll depth, signup intent, add-to-cart) and tie them to downstream value before you pour budget in. If CTR is great but downstream conversion lags, fix message match, page speed, and clarity. When the funnel hums, clone winning creatives across discovery partners, keep a rolling creative calendar, and treat cheap clicks as the start of a relationship — not the end goal.
Stop treating display as the only playground for cheap clicks. Connected TV, audio streams and podcasts, and in-game inventory are where attention is concentrated and banner fatigue is living its retirement. These formats give you time, tone, and context: long form visuals on the big screen, intimate voice-first moments in audio, and interactive placements inside gameplay that reward attention with real engagement.
Start CTV with tight creative and clear objectives. Produce 15 to 30 second spots designed for sound-on viewing, add subtitle-safe visuals, and cap frequency to avoid ad burnout. Buy both programmatic OTT and direct channel deals to control inventory and target via household or contextual segments. Measure completed view rate, cross-device lifts, and use incrementality tests rather than last-click to prove value before scaling.
With audio, play to the ear. Test dynamic ad insertion into podcasts and streaming, compare host-read reads with produced spots, and pair audio with companion in-app display to capture second-screen action. Track listen-through, session length changes, and branded search spikes. Use small genre buckets to find the sweet spot for voice, pacing, and call to action before rolling out broadly.
In-game ads reward creativity and timing. Deploy rewarded video, native overlays, and branded items in ways that enhance rather than interrupt gameplay. Sync messaging to game moments, A/B test CTA placement and reward levels, and measure downstream metrics like retention and lifetime value. Start with modest tests, scale threefold on clear winners, and reallocate budget from underperforming banners into these higher attention channels.
Cheap clicks hide in the repeat pool if you know where to look. Instead of chasing new audiences on the same crowded platforms, push high intent visitors back through the funnel with feed driven creative and time aware frequency. Tools like Criteo excel at product level dynamic ads that mirror the last thing a user viewed, while AdRoll makes sequential messaging and multi channel recovery simple. The key is not more impressions but right impressions at the right moment.
Start with tight recency buckets and progressive offers. Serve a soft reminder within 24 hours, a benefit driven creative at day three, and a heavy incentive only to visitors who hit product pages but did not convert by day seven. Use dynamic templates that pull price, image, and scarcity tags from the product feed so creative is always current. Test creative length, hero image, and CTA permutations to stop guessing and start compounding small wins.
Programmatic components matter for scale and cost control. Use automated bidding rules to optimize for incremental ROAS, add suppression lists to avoid wasting spend on converters, and seed lookalike pools from high lifetime value customers rather than one time buyers. Enable cross device stitching and server side events to close attribution gaps and feed the DSPs with higher quality signals. Consider dayparting and floor price caps so the algorithm buys cheap clicks when conversion probability is highest.
Quick tactical checklist to run this week: build 3 recency audiences, enable dynamic product templates, set ROAS based bidding and a suppression list for converters, and measure repeat revenue per cohort for 30 and 90 days. If CAC falls and repeat revenue climbs, scale; if not, tighten windows or swap creative. Small retargeting edits will often unlock the most defensible, low cost growth.
Aleksandr Dolgopolov, 21 November 2025