Broad Targeting vs Interest Stacks 2026: What Still Works After iOS
Broad targeting relies on minimal audience constraints and relies on algorithm optimization, while interest stacks layer multiple first-party or contextual signals to narrow reach - each trades scale for precision differently post-iOS.
Why This Matters Now
iOS 14.5 and subsequent privacy updates destroyed the targeting playbook most DTC operators built between 2015 and 2021. Third-party cookies, device IDs, and cross-site behavioral data evaporated. Platforms responded by rebuilding their ad systems around first-party data, algorithmic inference, and aggregate cohorts. Two competing strategies emerged: broad targeting, which minimizes audience definition and lets algorithms find buyers, and interest stacks, which layer first-party signals and contextual intent to pre-filter audiences.
The choice between them is not academic. It directly affects ROAS, CAC, and how much budget an operator can profitably deploy. A brand selling premium skincare faces a different calculus than a subscription box operator. Understanding which approach fits your funnel, product margin, and data infrastructure is the difference between scaling profitably and burning cash on unqualified clicks.
Broad Targeting: How It Works
Broad targeting means creating an audience with minimal constraints - often just age, geography, and platform (e.g., 'women 25-54 in the US on Instagram'). The platform's algorithm then uses conversion events from your pixel, CRM, or first-party audience to infer which users are most likely to convert. Meta's Advantage+ and Google's Performance Max are the canonical examples. The algorithm learns from your conversion data and optimizes delivery toward lookalike patterns within that broad pool.
The mechanic works because platforms have massive amounts of first-party data. Meta knows what users click, watch, search, and buy across its family of apps. Google knows search intent, YouTube watch history, and Gmail behavior. When an operator provides conversion signals (purchases, email signups, add-to-carts), the algorithm reverse-engineers the commonalities among converters and bids up inventory for similar users. No manual audience segmentation required.
Broad targeting excels at scale. Because the audience pool is large, the algorithm has more inventory to optimize against, more conversion signals to learn from, and more room to find edge cases and lookalikes. For operators with strong conversion tracking and sufficient daily conversion volume (typically 50+ conversions per day per campaign), broad targeting often delivers the lowest CAC and highest ROAS.
Interest Stacks: Precision Through Layering
Interest stacks are the opposite philosophy. Instead of relying on the algorithm to find buyers in a massive audience, operators manually or semi-manually define audiences by stacking multiple interest, behavioral, or contextual signals. A fitness brand might target 'users interested in CrossFit AND who follow fitness influencers AND visited a gym website in the last 30 days.' Each layer narrows the pool and increases the likelihood that remaining users are genuinely interested.
Interest stacks rely on first-party data (your CRM, pixel, or lookalike audiences) and contextual signals (keywords, publisher categories, app categories). They work because they reduce noise. A user who matches five intent signals is more likely to convert than a user who matches zero. The tradeoff is reach - a tightly stacked audience might be 10% the size of a broad audience, which means fewer impressions, fewer conversion signals for the algorithm to learn from, and potentially higher CPM (since you are competing for a smaller, more desirable pool).
Interest stacks are most effective for operators with smaller daily conversion volumes, high product margins, or complex customer profiles. A B2B SaaS company selling a $5,000 annual contract might run interest stacks because each conversion is valuable and the audience is inherently smaller. A DTC brand with a $30 AOV and thin margins might find interest stacks too restrictive.
Broad Targeting Post-iOS: What Changed
Before iOS 14.5, broad targeting was risky because third-party data was cheap and abundant. An operator could target 'women 25-54 interested in fashion' and get a reasonably qualified audience because the platform had granular behavioral data to filter within that group. Post-iOS, the algorithm has less granular data to work with, but it compensates by learning more aggressively from conversion signals.
The practical effect: broad targeting now requires more conversion volume to train effectively. Operators report that campaigns need 50-100 conversions per day (not per week) to stabilize. Below that threshold, the algorithm is essentially guessing. Additionally, broad targeting now performs better when conversion tracking is clean and first-party data is rich. If an operator's pixel is firing inconsistently or their CRM is not synced to the platform, broad targeting will underperform.
One counterintuitive benefit: broad targeting is now more resilient to iOS updates and platform changes. Because it relies on the platform's own first-party data and algorithmic inference rather than external data sources, it is less vulnerable to future privacy restrictions. Operators who switched to broad targeting in 2022-2023 have generally maintained performance, while those clinging to interest-based targeting have seen gradual degradation as third-party data sources dried up.
Interest Stacks Post-iOS: The Data Problem
Interest stacks hit a hard wall post-iOS because they depend on data that is now scarce or unavailable. Third-party interest categories (e.g., 'users interested in luxury goods') are less reliable. Behavioral targeting based on cross-site browsing is gone. Even first-party lookalike audiences are weaker because they are built on smaller, noisier datasets.
However, interest stacks have evolved. Operators now build stacks using contextual signals that survived iOS: search keywords, publisher categories, app categories, and first-party CRM data. A DTC brand can still stack 'users searching for product category keywords' with 'users who visited the website' with 'users in a lookalike audience from past buyers.' These stacks are narrower and noisier than pre-iOS equivalents, but they still work.
The key insight: interest stacks are now most effective when built on first-party data and intent signals, not third-party interests. An operator with a rich CRM, strong pixel implementation, and contextual targeting options can still build effective stacks. An operator relying on third-party interest categories will struggle. The data infrastructure matters more than ever.
Choosing Your Strategy: A Decision Framework
Start with conversion volume. If the brand generates 50+ conversions per day across all channels, broad targeting is the default. The algorithm has enough signal to learn effectively, and the reach advantage usually outweighs precision. If conversion volume is 10-50 per day, hybrid approaches work best - run broad campaigns alongside interest stack campaigns and let performance data decide. Below 10 conversions per day, interest stacks are usually necessary because the algorithm has too little signal to optimize a broad audience effectively.
Second, assess data infrastructure. Broad targeting requires clean conversion tracking, CRM sync, and ideally first-party audience uploads. If the brand's pixel is firing inconsistently or the CRM is siloed, broad targeting will disappoint. Interest stacks require rich first-party data (CRM, pixel, lookalikes) and access to contextual targeting options. If the brand has neither, neither strategy will work well - the problem is upstream.
Third, consider product economics. High-margin products (>50% gross margin) can absorb higher CAC and benefit from broad targeting's scale. Low-margin products (<30% gross margin) need precision and may require interest stacks to hit CAC targets. Subscription products with high LTV can afford to acquire customers at higher CAC upfront. One-time purchase products need immediate profitability.
Finally, test both. Run a broad targeting campaign and an interest stack campaign simultaneously for 2-4 weeks with equal budget. Measure ROAS, CAC, and conversion rate. The winner is usually clear. Many operators find that broad targeting wins on ROAS but interest stacks win on conversion rate - in that case, the decision depends on whether the brand is optimizing for efficiency or volume.
Practical Hybrid Approaches for 2026
Most successful DTC operators in 2026 are not choosing one strategy - they are layering both. A common pattern: run broad targeting campaigns for top-of-funnel awareness and scale, and run interest stack campaigns for retargeting and lookalike audiences. The broad campaign casts a wide net and generates conversion data. The interest stack campaign uses that data to build precise lookalike audiences and retarget engaged users.
Another approach: use interest stacks to segment the broad audience. Instead of one massive broad campaign, run three or four broad campaigns, each targeting a different interest stack (e.g., 'broad audience interested in fitness', 'broad audience interested in wellness', 'broad audience interested in health tech'). This gives the algorithm more focused conversion signals while maintaining reach.
A third pattern: start with interest stacks to validate product-market fit and CAC targets, then shift to broad targeting once conversion volume reaches threshold. Early-stage brands often lack the conversion volume for broad targeting to work, so they start with stacks. As they scale and conversion volume increases, they migrate to broad targeting to unlock efficiency gains.
FAQ
Does broad targeting still work without third-party data?
Yes, but it requires sufficient conversion volume and clean first-party data. Platforms like Meta and Google now rely on their own first-party data and algorithmic learning from conversion signals rather than third-party behavioral data. Broad targeting works best with 50+ daily conversions and accurate pixel implementation. Below that threshold, the algorithm lacks signal and performance degrades.
What is the minimum conversion volume needed for broad targeting?
Industry consensus is 50+ conversions per day per campaign, though some operators report success at 30-40 with very clean data. Below 10 conversions per day, broad targeting is unreliable because the algorithm cannot learn meaningful patterns. Between 10-50, hybrid approaches (broad + interest stacks) usually outperform pure broad targeting.
Are interest stacks dead post-iOS?
No, but they have evolved. Third-party interest categories are less reliable, but interest stacks built on first-party data (CRM, pixel, lookalikes) and contextual signals (keywords, publisher categories) still work. They are most effective for brands with high-margin products, small daily conversion volumes, or complex customer profiles where precision matters more than scale.
How do I know which strategy to test first?
Start with your conversion volume. If 50+ daily conversions, test broad targeting. If 10-50, run both simultaneously. If under 10, start with interest stacks. After 2-4 weeks, compare ROAS and CAC. The winner is usually clear, but many brands find they need both - broad for scale, stacks for precision and retargeting.
FAQ
Does broad targeting still work without third-party data?
Yes, but it requires sufficient conversion volume and clean first-party data. Platforms like Meta and Google now rely on their own first-party data and algorithmic learning from conversion signals rather than third-party behavioral data. Broad targeting works best with 50+ daily conversions and accurate pixel implementation. Below that threshold, the algorithm lacks signal and performance degrades.
What is the minimum conversion volume needed for broad targeting?
Industry consensus is 50+ conversions per day per campaign, though some operators report success at 30-40 with very clean data. Below 10 conversions per day, broad targeting is unreliable because the algorithm cannot learn meaningful patterns. Between 10-50, hybrid approaches (broad + interest stacks) usually outperform pure broad targeting.
Are interest stacks dead post-iOS?
No, but they have evolved. Third-party interest categories are less reliable, but interest stacks built on first-party data (CRM, pixel, lookalikes) and contextual signals (keywords, publisher categories) still work. They are most effective for brands with high-margin products, small daily conversion volumes, or complex customer profiles where precision matters more than scale.
How do I know which strategy to test first?
Start with your conversion volume. If 50+ daily conversions, test broad targeting. If 10-50, run both simultaneously. If under 10, start with interest stacks. After 2-4 weeks, compare ROAS and CAC. The winner is usually clear, but many brands find they need both - broad for scale, stacks for precision and retargeting.