Pricing is one of the most powerful growth levers for SaaS companies. Charge too little, and you limit revenue potential. Charge too much, and you choke off demand. The holy grail is finding the optimal price point that maximizes revenue while still attracting customers.
But how do you actually find that pricing sweet spot? One popular approach is the price split test, or A/B testing different prices. By offering the same product at different price points to similar audience segments, you can measure which price maximizes total revenue.
Split testing pricing can drive impressive results. HubSpot increased revenue from its Marketing Hub Starter product by 137% after testing three different price points. Litmus discovered an opportunity to nearly double average revenue per customer while maintaining conversion rates with a split test.
However, split testing pricing isn‘t a silver bullet. In fact, it can seriously backfire if not managed carefully. Showing different prices to different customers can breed perceptions of unfairness and damage your brand. Making decisions based on incomplete test data can lead you astray. And the operational complexity of running a multi-price test is often underestimated.
In this article, we‘ll walk through how to run an effective price split test while avoiding common pitfalls. We‘ll also explore some alternatives to split testing that can be just as impactful for optimizing SaaS pricing. Let‘s dive in.
What is Price Split Testing?
First, a quick definition. Price split testing (also called A/B testing prices) involves comparing two different prices for the same product to see which one performs better. Performance is usually measured in terms of total revenue, though other metrics like conversion rate, upgrades, and retention are also important.
In a typical price split test, website visitors are randomly divided into two groups, with each group shown a different price for the same product. For a SaaS company, this usually happens on the pricing page. One group might see a price of $49/month, while the other sees $99/month. The rest of the page content is identical.
Over the course of the test, the company tracks key metrics like visits, sign-ups, and revenue for each group. If the $99 price yields more revenue in total, the company would likely choose that as the winner. Price split testing aims to find the price that strikes the optimal balance between volume (number of customers) and revenue per customer.
Benefits of Price Split Testing
Why bother with price split testing? After all, it takes time and resources to execute properly. But the potential upside is huge:
-
Increased revenue and profitability. Finding the optimal price point allows you to capture more value from each customer. Even small changes in price can have a big impact on the bottom line.
-
Better understanding of customer price sensitivity. Split testing reveals how different prices impact demand for your product. This is powerful data that can inform packaging, discounting, and marketing as well as pricing.
-
Ability to target different segments. Customers are not one homogenous group. Split testing can uncover different price sensitivities among key segments that can be targeted accordingly.
-
Faster iterations and optimization. Compared to one-time pricing changes, split testing allows for ongoing experiments and faster reaction to the market. Pricing can be optimized gradually over time.
Of course, these benefits only materialize if the price test is set up and interpreted properly. Let‘s look at some best practices for an effective split test.
How to Split Test SaaS Pricing: A Step-by-Step Guide
Effective price split testing is both an art and a science. Here‘s a proven step-by-step process to get it right:
-
Choose the right products to test. Focus your test on a specific product or tier rather than your whole pricing matrix. This isolates price as the variable and avoids confounding factors. If testing multiple products, choose ones that are similar in features and value proposition.
-
Determine price points to test. Look at competitive benchmarks, customer research, and historical pricing data to define the range of prices to test. Most A/B testing tools can only compare two variants at a time, so limit your initial test to 2-3 price points at most. Make the price differences between variants significant enough to measure an effect (aim for a 20-30% difference).
-
Calculate your target sample size. To reach statistically significant results, you‘ll need a minimum number of visitors to see each price variant. Use an A/B test significance calculator to determine your sample size based on current conversion rates and the desired level of improvement. For low-traffic sites, you may need to run the test for a month or more.
-
Set up the test in your A/B testing tool. Most major A/B testing tools like Optimizely and VWO have a visual editor that allows you to change prices without coding. Be sure to QA the variations thoroughly. To maintain consistency, consider saving the visitor‘s price variant in a cookie so they see the same price on subsequent visits.
-
Launch the test and wait for significance. Let the test run until you‘ve reached the predetermined sample size and statistical significance (usually a 95% confidence level). Avoid peeking at the results or making decisions prematurely. Remember that statistical significance doesn‘t always equal practical significance – a 5% lift on a $10,000 MRR test variant may not be worth the tradeoffs.
-
Measure revenue and key behavioral metrics. Conversion rate alone doesn‘t tell the whole story. Revenue per visitor is the gold standard for price tests – whichever variant drives more total revenue wins. Also look at key metrics like average revenue per user, retention, upgrades, and support costs to get a complete picture.
-
Segment the results by customer attributes. Slice the data by key segments like company size, industry, and use case. You might find that higher prices work better for enterprise customers while lower prices drive more revenue from SMBs. This can reveal opportunities for segmented pricing.
-
Iterate and retest based on insights. A single test is rarely conclusive. Analyze the results and form a new hypothesis to test. For example, if the lower-priced variant won, try testing an even lower price. If you saw a wide gap between variants, test a price in the middle. Continuously iterate to hone in on the optimal price.
Here‘s an example of how multi-step split testing led to a winning price for SaaS company Appcues:
| Test | Control Price | Variant Price | Result |
|---|---|---|---|
| 1 | $79/mo | $99/mo | 13% lift in ARPU, flat conversion |
| 2 | $99/mo | $129/mo | 20% drop in conversion, flat revenue |
| 3 | $99/mo | $89/mo | 5% lift in conversion and revenue |
Through three rounds of testing, Appcues zeroed in on $89/month as the optimal price for its core product – a sweet spot between volume and revenue per user. This kind of iterative approach is the key to successful price optimization.
The Risks of Split Testing SaaS Pricing
While the potential rewards of price split testing are significant, there are risks to be aware of:
-
Customer perception of unfairness: If customers become aware that others are paying different prices for the same product, they may feel cheated. The appearance of a "price cut" for some can damage trust in your brand. In a worst case scenario, it can even draw legal scrutiny around discriminatory pricing. To mitigate this, don‘t show radically different prices to logged-in customers vs. first-time visitors. Consider excluding extremely high-value customers from tests altogether.
-
Operational complexity: Running a split test means you‘ll have customers on different price points, which can cause headaches in billing, feature gating, and support. You‘ll need to update your systems to handle multiple active price points for the same product. Think through the operational implications before launching a test and have a plan to migrate "losers" to the new price smoothly.
-
Statistical significance: Reaching a 95% confidence level in a SaaS price test can be challenging, especially for lower-volume products. You may need to run a test for several months to reach significance. Even then, the results may not be practically meaningful. Be wary of making decisions based on small sample sizes or weak effects. When in doubt, err on the side of caution and don‘t make changes unless you have strong data to support it.
-
Strategic considerations: Pricing is not just a numbers game. There are qualitative and strategic factors to weigh beyond the raw test data. How will a price change impact your market positioning? Your sales process? Your long-term product roadmap? Don‘t let short-term price testing override your overarching business strategy. Sometimes it may be better to leave money on the table in the near term to achieve a larger long-term goal.
A real-world example illustrates these risks. In 2000, Amazon offered DVDs at different prices to different customers as a price test. When customers found out, there was a major backlash. Amazon refunded the price differences and publicly apologized. The PR hit likely wiped out any revenue gains from the test. While technology has evolved since then to allow cleaner split tests, the fundamental risks remain.
Alternatives to Split Testing Pricing
Given the risks and limitations of split testing pricing, it‘s worth considering alternative approaches:
1. Test pricing page design and positioning
Instead of testing the price itself, try testing different ways of presenting and framing your pricing. This could include the design and layout of the pricing page, the naming and descriptions of different plans, and the overall value proposition. By optimizing how you sell your product, you may be able to charge a higher price without changing the number itself.
2. Survey customers on willingness to pay
Asking customers directly about their price sensitivity can yield useful insights without the risk of a live split test. Use surveys, interviews, or user research to understand what customers value in your product and how much they‘d be willing to pay for it. Focus on relative value – e.g., "How much more would you pay for feature X?" – rather than absolute numbers.
3. Monitor and analyze post-purchase behavior
Your existing customer base contains a wealth of pricing data. Analyze metrics like feature usage, retention, upsells, and support costs across different customer segments. You may find that certain types of customers are getting outsized value relative to what they‘re paying. This can inform targeted price increases or new premium offerings.
4. Launch new pricing in phases
Instead of an A/B test, roll out pricing changes gradually to mitigate risk. Start with a small segment of users, measure results, and iterate before expanding to your full user base. This allows you to test new pricing in a controlled way and makes it easier to roll back if needed. You can also use a gradual ramp-up to manage the perception of a price increase.
Here‘s a summary of the pros and cons of each approach:
| Method | Pros | Cons |
|---|---|---|
| Split testing | High volume of data, isolates price as variable | Unfairness to customers, operational complexity, significance challenges |
| Pricing page optimization | Lower risk, focuses on value over cost | May not reveal true willingness to pay, smaller upside |
| Customer surveys | Direct input from buyers, informs segmentation | Self-reported data may be unreliable, small sample sizes |
| Behavioral analysis | Uses real purchase data, informs packaging | Only looks at existing customers, harder to isolate pricing impact |
| Phased rollouts | Lower risk, easier to manage perception | Slower to generate learnings, potential for customer confusion |
Ultimately, the best approach depends on your specific product, market, and growth stage. Many SaaS companies use a combination of methods to continuously optimize pricing. The key is to always be testing and iterating based on data.
Translating Pricing Test Results Into Action
Collecting pricing data is only half the battle. The real challenge is interpreting the results and translating them into action. Here are some tips for making smart decisions based on pricing test results:
- Combine quantitative and qualitative insights. Don‘t just rely on the hard numbers from split tests or surveys. Layer in qualitative feedback from customers, prospects, and internal teams to get a complete picture. Sometimes a "failed" test can still reveal valuable insights that inform your pricing strategy.
- Make revenue the key metric. It‘s easy to focus on conversion rate or churn, but revenue is the ultimate goal of pricing optimization. Look at metrics like MRR, TCV, and LTV by segment to understand the full impact of pricing changes. Be willing to trade off short-term conversions for long-term revenue gains.
- Model out scenarios. Before making a pricing change, model out the expected impact on revenue and other key metrics. This will help you set realistic expectations and avoid surprises. Use your test data to inform the model, but don‘t treat it as gospel. Leave room for error and consider multiple scenarios.
- Get buy-in from key stakeholders. Pricing changes can have far-reaching impacts across marketing, sales, product, and finance. Make sure all key stakeholders are aligned on the rationale and expected outcomes before rolling out a new pricing model. Use your test results to build a compelling case for change.
- Communicate changes clearly to customers. How you roll out pricing changes is just as important as the changes themselves. Be transparent with customers about why you‘re making a change and how it will benefit them in the long run. Consider grandfathering existing customers or offering them incentives to move to the new pricing.
Pricing strategy is a high-stakes game for SaaS companies. The right price can supercharge your growth, while the wrong price can stunt it. Split testing is a powerful tool for optimizing your pricing, but it‘s not without risks and limitations. By following the best practices laid out in this guide and considering alternative approaches, you can find the pricing sweet spot that maximizes revenue and aligns with your broader business goals.
Remember, pricing is an ongoing process, not a one-time event. As your product and market evolve, your pricing strategy should evolve with it. The most successful SaaS companies are relentless about testing, iterating, and adapting their pricing based on data and customer feedback. By making pricing optimization a core competency, you can unlock the full revenue potential of your business.
){kind=link}