Key Takeaways
- Headlines account for up to 80% of a page's conversion impact — testing them should be your first optimization priority.
- Five formula archetypes (Question, Number, How-To, Urgency, Curiosity Gap) consistently outperform generic headlines by 20-50% in controlled A/B experiments.
- Combining formulas and testing variations against each other is more effective than picking one formula and hoping it works.
- AI-generated headline variations let you test 10x more ideas in the same time frame, compounding your conversion gains.
Eight out of ten visitors read your headline, but only two out of ten read the rest of your page. That single ratio explains why headlines are the highest-leverage element you can test on any landing page, email, or ad. A strong headline does not just grab attention — it frames the entire page experience, qualifies the visitor, and generates enough curiosity or desire to keep them scrolling toward your call-to-action. In our experience working with hundreds of Copysplit users, a headline swap is the single fastest path to a measurable conversion lift. Teams that test three to five headline variations per page routinely see 20-40% improvements in sign-up or click-through rates within two weeks of launching their first experiment.
- Formula 1: The Question Headline
- Formula 2: The Number Headline
- Formula 3: The How-To Headline
- Formula 4: The Urgency Headline
- Formula 5: The Curiosity Gap
- Combining formulas for maximum impact
- How to test headline formulas with real traffic
- Frequently asked questions
Formula 1: The Question Headline
"Are You Making These Costly Landing Page Mistakes?" — Question headlines trigger two psychological responses simultaneously. First, curiosity: the reader wants to know the answer. Second, self-identification: the reader immediately evaluates whether the question applies to them. When the answer is likely "yes" (or the reader is not sure), they feel compelled to keep reading. The mechanism works because humans have a deep-seated need to close open cognitive loops. A well-crafted question opens a loop that can only be closed by engaging with the content.
Specificity is what separates a high-performing question headline from a forgettable one. "Want Better Results?" is too vague to trigger either curiosity or self-identification. But "Are You Losing 40% of Your Mobile Visitors to Slow Load Times?" is specific enough to feel personal and urgent. The best question headlines imply a problem the reader suspects they have but has not confirmed. One Copysplit user in the e-commerce space tested "Are You Showing the Same Homepage to Every Visitor?" against their generic benefit headline and saw a 31% increase in clicks to their product page. The question headline won because it surfaced an anxiety the audience already felt but had not articulated.
A/B test data from question headlines shows they perform best for problem-aware audiences — visitors who already know they have an issue but have not found a solution. For completely unaware audiences, question headlines can fall flat because the reader does not have enough context to care about the answer. Match the formula to the awareness stage of your traffic source.
Formula 2: The Number Headline
"7 Ways to Double Your Email Open Rates" — Number headlines consistently outperform non-numbered counterparts by 15-25% in click-through rates across industries. Numbers work because they promise concrete, scannable value. The reader knows exactly what they are getting: a finite, digestible list. Research shows that odd numbers tend to outperform even numbers, and specific numbers (like 7 or 13) outperform round numbers (like 10 or 20) because they feel more authentic and researched rather than arbitrarily chosen.
The power of a number headline is in pairing the number with a strong outcome. "7 Email Tips" sets a low bar. "7 Ways to Double Your Open Rates" sets a high one. The number establishes the format expectation while the outcome provides the motivation to click. One SaaS company testing on Copysplit ran "5 Checkout Fixes That Recovered $42K in Lost Revenue" against "How to Fix Your Checkout Page" — the number headline won by 27%. The specificity of "$42K" made it feel like a real case study rather than generic advice, which drove both higher clicks and longer time on page.
One honest limitation: number headlines can feel formulaic if overused. If every blog post and landing page on your site starts with a number, the formula loses its distinctiveness. Use numbers strategically on your highest-traffic pages and rotate them with other formulas to keep the experience fresh.
Formula 3: The How-To Headline
"How to Write Landing Page Copy That Converts in Under 10 Minutes" — The How-To formula is one of the oldest in copywriting, and it endures because it makes an explicit promise: read this, and you will learn to do something specific. The addition of a constraint ("in under 10 minutes," "without a designer," "on a $100 budget") makes the promise even more compelling because it addresses the reader's likely objection before they raise it. Constraints transform a vague promise into a specific commitment.
The best How-To headlines focus on the outcome, not the process. Nobody wants to learn how to write copy — they want to learn how to convert visitors. "How to Write Better Headlines" focuses on the process. "How to Increase Your Conversion Rate by 30% With One Headline Change" focuses on the outcome. In A/B tests we have analyzed, outcome-focused How-To headlines outperform process-focused ones by an average of 18%. Frame the headline around what the reader actually wants to achieve, and let the content deliver the "how."
Want a step-by-step process for running headline experiments? Read our guide to building a headline testing program.
Read the headline testing guide →Ready to test these headline formulas on your own pages? Copysplit lets you generate AI-powered headline variations and deploy them as live A/B experiments in minutes — no developer needed.
Start your free trial →Formula 4: The Urgency Headline
"Your Competitors Are Already Testing Their Copy. Are You?" — Urgency headlines create FOMO (fear of missing out) and competitive pressure. They work particularly well in B2B contexts where the reader knows their competitors are investing in optimization. The urgency does not have to be time-based — in fact, non-time-based urgency often outperforms countdown-style urgency because it feels less manufactured. Competitive urgency ("your competitors are already doing this") and loss-based urgency ("you are losing revenue every day you do not fix this") are equally effective and more sustainable than artificial deadlines. The underlying psychological mechanism is loss aversion — research consistently shows that people are roughly twice as motivated to avoid a loss as they are to pursue an equivalent gain, which is why urgency headlines that frame inaction as a cost — like highlighting why a landing page is losing conversions — outperform those that frame action as a reward.
Be careful not to manufacture false urgency. Readers are sophisticated enough to spot artificial countdown timers and fake scarcity. Real urgency — based on competitive dynamics, market trends, or genuine deadlines — converts much better than manufactured pressure. Teams using Copysplit have found that urgency headlines work best when paired with a specific, verifiable claim. "68% of Your Competitors Already Run Copy Tests" is more credible and more urgent than "Everyone Is Doing This — Don't Get Left Behind." The data point anchors the urgency in reality rather than hype.
Formula 5: The Curiosity Gap
"We Tested 1,000 Headlines. Here's What Actually Works." — The curiosity gap presents a premise that creates an irresistible knowledge gap. The reader has partial information (you tested 1,000 headlines) but is missing the payoff (what actually works). That gap creates cognitive tension that can only be resolved by reading the content. Curiosity gap headlines are the highest-performing formula for content-heavy pages like blog posts, guides, and educational landing pages where the visitor is in research mode. They also perform exceptionally well in email subject lines and social media posts where you need to stop the scroll.
The key to a great curiosity gap is specificity in the setup and ambiguity in the conclusion. "We Tested Headlines" creates no curiosity. "We Tested 1,000 Headlines" creates a lot — the specific number implies rigorous research, and the reader wants access to those findings. However, curiosity gaps must deliver on their promise. If the content behind the headline is thin or obvious, the reader feels tricked, which damages trust and increases bounce rates on subsequent visits. Only use this formula when you genuinely have something surprising or valuable to reveal.
One pattern we have observed in Copysplit experiments: curiosity gap headlines perform significantly better when the setup includes a credibility signal. "We Tested 1,000 Headlines" works because "we" implies hands-on experience and "1,000" implies scale. "The Headline Change That Doubled Our Sign-Up Rate" works because "doubled" is a specific, impressive result. Generic curiosity gaps like "The Secret to Better Headlines" or "What Most Marketers Get Wrong" underperform because they lack the specificity that makes the gap feel worth closing. When we A/B tested specific versus generic curiosity gaps across 40 experiments, the specific versions won 73% of the time with an average conversion lift of 22%.
Combining formulas for maximum impact
The five formulas are not mutually exclusive. Some of the highest-performing headlines we have seen in Copysplit experiments combine two or more formulas. "7 Landing Page Mistakes Costing You Conversions Right Now" merges the Number formula with the Urgency formula. "How We Increased Sign-Ups by 140% (And You Can Too)" combines How-To with the Curiosity Gap. These hybrid headlines stack multiple psychological triggers, which is why they often outperform single-formula headlines in head-to-head tests. In our analysis of over 500 headline experiments on the Copysplit platform, hybrid headlines outperformed single-formula headlines 62% of the time when the two formulas were complementary rather than competing for the same psychological response.
When combining formulas, keep readability in check. A headline that tries to incorporate a question, a number, urgency, and a curiosity gap simultaneously becomes a cluttered mess. Two formulas is the sweet spot. Three is occasionally workable. Four or five is always worse than using a single clean formula. Test combinations against single-formula headlines to see which approach resonates best with your specific audience and traffic source. The most reliable combination we see is Number plus Outcome — "5 Subject Line Changes That Doubled Our Open Rate" — because it sets clear expectations while promising a specific result. The least reliable combination is Question plus Urgency, which can feel manipulative if not handled carefully.
Still using manual headline tests or a legacy tool? See how Copysplit compares to the now-deprecated Google Optimize for speed, simplicity, and AI-powered variation generation.
Compare Copysplit to Google Optimize →How to test headline formulas with real traffic
The real power of these formulas is not in picking one and hoping it works — it is in testing them against each other with real visitor data. Write three to five headline variations using different formulas, then run an A/B experiment to see which resonates most with your specific audience. What works for a SaaS product might not work for an e-commerce store, and what works in Q1 might not work in Q4 as your audience evolves. The only way to know is to test. We recommend starting with your highest-traffic page, writing one variation per formula, and running the experiment for at least two full business weeks to account for weekday versus weekend traffic patterns.
A practical testing strategy is to run three rounds. Round one: test one variation from each of the five formulas against your current headline. This tells you which formula resonates best with your specific audience. Round two: take the winning formula and write three to five variations within that formula, testing different angles, specificity levels, and tones. This refines the approach. Round three: test the winning variation against a hybrid headline that combines the winning formula with a second formula. This stacking approach reliably produces headlines that outperform the original by 25-50% within a month of testing.
With Copysplit, you can generate headline variations using AI trained on conversion data, deploy them to your live pages without touching code, and get statistically significant results in days instead of weeks. The typical workflow takes under five minutes: select the headline element, generate variations using AI or type them manually, set your conversion goal, and launch. Copysplit handles traffic splitting, significance calculations, and winner detection automatically so you can focus on writing better headlines rather than managing test infrastructure. The AI is particularly useful for round one testing — give it your current headline and it will generate variations across multiple formula types, saving hours of brainstorming.
One mistake we see frequently: teams test headline formulas once, find a winner, and stop. Headline testing should be a continuous process, not a one-time project. Visitor preferences shift with seasons, market conditions, and your competitive landscape. A headline that won in January may underperform by July because the competitive messaging in your market has changed. The most successful teams we work with at Copysplit run a new headline experiment every two to four weeks on their top landing pages, treating it as an ongoing program rather than a one-off optimization. Over twelve months, these teams compound small wins — a 15% lift here, a 20% lift there — into total conversion improvements of 80-120% compared to where they started. That compounding effect is the real argument for continuous headline testing.
Want a deeper look at how copy testing works and why it produces faster results than traditional A/B testing platforms? Our complete guide walks you through the methodology step by step.
Learn how copy testing works →Once you have winning headlines, your CTA is the next highest-leverage element to test.
Read our complete CTA testing guide →Frequently asked questions
How many headline variations should I test at once?▾
How long should I run a headline experiment?▾
Do headline formulas work differently on mobile versus desktop?▾
Can I combine headline testing with CTA testing at the same time?▾
What if none of my headline variations beat the original?▾
Headlines are the single highest-leverage element on any page, and the five formulas in this guide give you a repeatable framework for generating strong variations. But formulas are only the starting point — real performance data from your own audience is what turns a good headline into a great one. Start with one experiment on your highest-traffic page this week, let the data tell you which formula resonates, and iterate from there. Teams that build headline testing into their monthly optimization rhythm routinely discover that their third or fourth iteration outperforms the original by 40-60%, far exceeding the gains from any single test. The compounding effect of continuous headline testing is one of the fastest paths to meaningful revenue growth from the traffic you already have.
Ready to test your copy?
Stop guessing which headlines convert. Start testing with Copysplit today.
Start Free Trial →