# Hypothesis Library — Ad Creative Testing Pre-built hypothesis templates organized by test type. Use as starting points and customize with your specific product, audience, and baseline metrics. --- ## Hook Tests ### Hook Type — Problem vs. Product **Hypothesis:** > If we change the opening hook from a product demonstration shot to a pain-point question ("Struggling with [problem]?"), then we expect a [15–25]% improvement in thumb-stop rate (3-second view rate) and a [10–20]% reduction in CPA, because problem-aware audiences respond more to problem framing than product framing in the early stages of purchase consideration. **When to use:** Product in a competitive category where buyers are aware of the problem but comparing solutions. --- ### Hook Type — Creator vs. Brand **Hypothesis:** > If we change the hook from a polished brand video opening to a creator-style direct-to-camera opening, then we expect a [20–35]% improvement in video completion rate and a [15–25]% improvement in conversion rate from video viewers, because native-looking content on TikTok has lower skip rates and higher trust signals than brand-produced content. **When to use:** TikTok or Instagram Reels campaigns where your current creative looks obviously like an ad. --- ### Hook Type — Social Proof vs. Feature **Hypothesis:** > If we change the hook from leading with a product feature to leading with a customer result or testimonial, then we expect a [10–20]% increase in landing page click-through rate, because social proof is a stronger initial trust signal than feature claims for first-time buyers in this category. **When to use:** Product in health, beauty, fitness, or any category where results-based claims build more immediate credibility than technical features. --- ### Hook Type — Urgency vs. Benefit **Hypothesis:** > If we change the hook from a benefit statement ("Get clearer skin in 7 days") to a scarcity/urgency frame ("Only 3 days left at this price"), then we expect a [5–15]% increase in CTR but potentially flat or decreased conversion rate — we are testing whether urgency drives action without degrading purchase intent quality. **When to use:** Testing urgency carefully; monitor conversion rate and AOV, not just CTR. --- ## Offer Tests ### Offer Structure — Discount vs. Bonus **Hypothesis:** > If we change the offer from a 20% price discount to a "buy one get one free" bundle at the same effective cost to the company, then we expect a [10–20]% increase in conversion rate at equivalent margin, because BOGO framing creates perceived higher value than a percentage discount for this product category. **When to use:** When you want to protect price perception while offering equivalent incentive value. --- ### Offer Structure — Flat Discount vs. Free Shipping **Hypothesis:** > If we change the offer from "£5 off your first order" to "free shipping on your first order," then we expect equivalent or higher conversion rate at a lower actual cost (since free shipping costs us less than £5 discount), because free shipping removes a decision friction point without anchoring to a lower price point. **When to use:** Products where shipping cost is a visible friction point at checkout. --- ### Offer Structure — Single Unit vs. Bundle **Hypothesis:** > If we present the primary offer as a 3-unit bundle instead of a single unit (at proportionally lower per-unit price), then we expect lower conversion rate but higher AOV and higher gross margin per order, because bundle framing reaches buyers with higher purchase intent and improves unit economics even at lower conversion volume. **When to use:** Products with repeat purchase intent and reasonable shelf life or stacking utility. --- ## Landing Page Tests ### Page Type — Dedicated PLP vs. Homepage **Hypothesis:** > If we send ad traffic to a dedicated product landing page instead of the store homepage, then we expect a [30–60]% improvement in landing page conversion rate, because removing navigation distractions and maintaining creative-to-page message match reduces decision friction for buyers arriving with high purchase intent. **When to use:** Almost always valid for paid traffic; homepage as a landing page is almost always weaker than a dedicated page. --- ### Page Element — Reviews Placement **Hypothesis:** > If we move customer reviews to above the fold on the product landing page (currently below fold), then we expect a [10–20]% improvement in conversion rate, because social proof at the decision point reduces hesitation before the buyer has to scroll to find it. **When to use:** Pages where reviews exist but are buried; especially for higher-ticket products. --- ### Page Element — Video vs. Static Hero **Hypothesis:** > If we replace the static hero image with a 15-second product demonstration video as the hero element, then we expect a [15–30]% improvement in landing page conversion rate for mobile visitors, because video communicates product function and credibility faster than static imagery for tactile or functional products. **When to use:** Products that are difficult to understand from a photo alone (supplements, tools, skincare with visible results). --- ## Audience Tests ### Audience Type — Broad vs. Interest Targeting **Hypothesis:** > If we run Variant B to a broad audience (age/gender only, no interest targeting) while Variant A runs to our current interest-targeted audience, then we expect Variant B to achieve lower CPA after the learning phase due to the platform's algorithm having more signals to find optimal buyers, because recent platform data suggests broad targeting outperforms interest targeting for proven creatives with 500+ conversion history. **When to use:** Campaigns with established conversion data ready to scale; broad testing is less appropriate for new products or low-volume campaigns. --- ### Audience Type — Retargeting Window **Hypothesis:** > If we change our retargeting window from 30-day visitors to 7-day visitors (higher intent), then we expect a [20–40]% reduction in CPA for retargeting campaigns, because 7-day visitors have higher purchase intent than 30-day visitors and reducing audience dilution improves ad relevance. **When to use:** Retargeting campaigns with sufficient 7-day audience size (minimum 1,000+ visitors/week). --- ## Format Tests ### Format — 15s vs. 30s Video **Hypothesis:** > If we change our ad format from 30-second video to 15-second video for the same creative concept, then we expect equal or higher conversion rate at lower CPM, because shorter videos have higher completion rates and TikTok/Meta may serve 15s content more efficiently, while 30s adds runtime that may not contribute to conversion for this offer type. **When to use:** When testing whether your current video length is adding value or adding friction. --- ### Format — Video vs. Static Image **Hypothesis:** > If we test a high-quality static image ad alongside our current video creative (same offer, same audience), then we expect the static image to achieve a lower CPM and possibly competitive CPA, because static images often have lower production cost and the algorithm delivers them efficiently for purchase-intent campaigns in mature stages. **When to use:** Scaling phase testing when you want to diversify creative formats and reduce production dependency on video. --- ## Quick-Reference — Which Variable to Test First Use this sequence for a new product or campaign: 1. **Hook first** — Hook performance has the highest leverage on downstream metrics. Get this right before testing anything else. 2. **Offer second** — Once you have a hook that performs, test whether the offer structure can improve conversion rate. 3. **Landing page third** — With a proven hook and offer, test landing page optimizations to squeeze the conversion rate. 4. **Audience fourth** — With proven creative, test whether audience expansion or refinement improves CPA at scale. 5. **Format last** — Format tests are useful for efficiency at scale, not for finding fundamental improvements.