/
Pricing & Monetization
A "Most Popular" badge is not a prettier card, it's a choice-architecture intervention. Done well, it makes the best-fit plan easier to recognize. Done badly, it makes the plan the company wants sold harder to refuse.
what would you do?
Your pricing page shows three plans with the middle tier badged "Most Popular." A stakeholder wants to also add a bigger badge, a colored background, an oversized card, and "BEST DEAL!!!" copy to push conversion higher. What should you actually do first?
Pick the highest-impact fix:
A
Add all the requested visual elements at once so the plan is impossible to miss
B
Remove the badge entirely since highlighting is manipulative
C
Copy Loom's exact visual treatment since it's a well-known SaaS pricing page
D
Verify the "Most Popular" claim is actually true for that plan, and if you want to test the visual treatment, isolate one variable at a time
The Problem
Pricing pages force several decisions into a few seconds. The visitor has to understand the value metric, compare features, infer which customer type each tier represents, interpret price units, choose a billing cadence, estimate future usage, and decide whether the additional value of the next tier justifies the additional commitment. Stripe's SaaS pricing guidance emphasizes that packaging should make it easy for buyers to identify the tier appropriate to them, because confusing tier structures can slow the decision or push customers into an inappropriate package [6].
This creates a fundamental tension: the customer wants confidence, fit, transparency, and low regret, while the business wants conversion, a desirable plan mix, annual commitments, expansion potential, and sustainable unit economics. The pricing interface has to reconcile both.
Pricing success ≠ maximum clicks on the highlighted card. Pricing success = a function of conversion, plan mix, revenue, retention, and trust together.
Classic research on anchoring, asymmetric dominance, compromise effects, and defaults shows that the structure and starting point of a choice can influence decisions even when the underlying options remain unchanged [1]. But there is no credible universal benchmark such as "highlighting the middle plan increases conversion by 20%." GoodUI documents a Netflix experiment in which a balanced three-plan page was challenged by a preselected single mid-tier recommendation, and the recommendation was rejected [3]. A recommendation that moves customers into an expensive tier they later downgrade or cancel can look brilliant in a seven-day conversion dashboard and terrible in a twelve-month cohort.
The ethical boundary matters just as much as conversion. The European Commission explicitly describes "false hierarchy" as an interface that steers consumers through visual design or language, and its 2022 dark-pattern sweep found 54 of 399 checked websites directing consumers toward certain choices through interface design or wording. The FTC similarly identifies buried key terms, deceptive subscription flows, preselected choices, and interfaces that impair informed choice as dark-pattern concerns [5].
The Design Principle
The strongest implementation is not "make the middle plan bigger." It is: identify the plan that genuinely fits the largest valuable segment, explain why it fits, create one clear visual focal point, and test the entire decision system against revenue and retention outcomes [12].
Choose the label according to what you can substantiate
LABEL | WHAT THE USER INFERS | EVIDENCE YOU SHOULD HAVE |
|---|---|---|
"Most Popular" | More customers choose this plan than alternatives | Current plan-selection data for the relevant population |
"Recommended" | The company believes this is the strongest general fit | A defensible target persona or use-case rationale |
"Best Value" | This option provides unusually favorable value for money | A transparent comparison of benefits and economic trade-offs |
"Recommended for you" | The recommendation reflects information about this visitor | Actual segmentation or personalization inputs |
"Save 20% annually" | Annual billing is economically cheaper than monthly | Accurate monthly-equivalent and full-term arithmetic |
Conflating these labels can make the page more persuasive in the short run while reducing credibility [13]. Build one visual focal point, a badge, border or background differentiation, a slightly more prominent CTA, but the highlighted card should not be the only one with a usable CTA, legible price, or sufficient feature information [14]. A useful hierarchy: recommendation badge, plan name and who it's for, price and billing unit, primary CTA, three to five differentiating benefits, then detailed comparison. And the ethical test is simple: would the design still feel fair if the customer understood exactly why this plan was highlighted? If not, the intervention is probably optimizing against the customer rather than for them.
Real Examples
A Dark Ribbon Making an Economic Claim
"Best value" attached to the plan with the most complete feature list
Superhuman
The design decision
Superhuman's pricing page attaches a solid dark "Best value" ribbon to its $33/month Business plan, sitting between Pro at $12 and a custom Enterprise tier. Each card states who it's for in one line, "For growing businesses", and lists every feature carried over from the tier below before adding what's new, so the Business column visibly contains the most complete stack of AI tools, Mail, Grammarly, Coda, and Go, plus business-specific capabilities.
Why it works
"Best value" is an economic assertion, and Superhuman substantiates it by making the accumulated feature list visible rather than just asserting the label. The audience line under each plan name, "For individuals," "For professionals and small teams," "For growing businesses," lets a visitor self-select before ever reaching the price, which is exactly the segment-first approach this guide's framework recommends.
A Popularity Badge With a Colored Border Focal Point
One card visually separated from its neighbors by color, not just size
A Neutral Comparison With No Highlight at All
Color-coded plan names replace a badge, letting the feature grid do the work
Match the Highlight Type to the Claim
HIGHLIGHTING TYPE | CORE MESSAGE | BEST FIT | PRINCIPAL RISK |
|---|---|---|---|
Popularity cue | "Other customers tend to choose this" | Broad self-serve market where one tier genuinely dominates | Social-proof claim becomes misleading if usage data doesn't support it |
Editorial recommendation | "We believe this is the strongest mainstream fit" | Products with recognizable customer maturity stages | Company economics can masquerade as customer fit |
Value recommendation | "This gives the strongest price/value balance" | Plans where incremental benefits can be compared clearly | "Best" is vague or deceptive if the basis is invisible |
Personalized recommendation | "Based on your context, start here" | Logged-in users, known team size, known usage | False personalization or discriminatory pricing logic |
Billing-cadence nudge | "Commit longer and pay less per month" | Products with good retention and predictable recurring use | Hiding annual charge, renewal terms, or discount arithmetic |
Neutral comparison | "Choose based on your organization's requirements" | Complex B2B products where one recommendation would be misleading | More comparison effort if differentiation is weak |
Mistakes That Kill Success
avoid this
Turning Every Visual Knob to Maximum
A colored background, oversized card, giant badge, larger price, animated border, different CTA color, and "BEST DEAL!!!" copy can make the page feel like an advertisement rather than a decision tool [14].
Fix
Make one or two visual dimensions dominant and keep the other plans fully usable, with a legible price, working CTA, and enough feature information.
avoid this
Assuming the Middle Plan Should Always Win
Compromise-effect research explains why a middle alternative can become attractive, but SaaS customers arrive with heterogeneous needs. GoodUI's Netflix example is instructive: a single preselected mid-tier recommendation was tested against a traditional balanced comparison and was rejected [32].
Fix
Treat "highlight the middle" as a testable hypothesis, not a law. Identify the plan that fits your largest valuable segment first.
avoid this
Using an Annual Monthly-Equivalent as the Displayed Amount
"$29/month" can create a much softer psychological price than "$348 charged today," even though both describe the same annual contract. FTC guidance specifically calls out buried material terms and deceptive subscription design [13].
Fix
Place the billing period, total charge, renewal cadence, and saving calculation directly beside the visible price or CTA.
Metrics That Matter
Attention and Selection Metrics
A highlighted plan should be judged at three levels: attention and selection, commercial conversion, and downstream quality. Stripe recommends examining conversion, ARPU, revenue, churn, and segment-level performance together [4].
Core Formulas
Pricing-page paid conversion = new paid customers / eligible unique pricing visitors
Highlighted-plan take rate = highlighted-plan purchases / all paid-plan purchases
Plan-mix shift = share(variant) − share(control)
Relative conversion lift = (CR(variant) − CR(control)) / CR(control) × 100%
Revenue and Guardrail Metrics
For most self-serve subscription businesses, revenue or contribution margin per eligible pricing visitor is more informative than highlighted-plan take rate alone. A badge that shifts everyone into the target plan while reducing total checkout conversion can be a commercial loss [4].
Downstream Quality Guardrails
At minimum, monitor refund rate, early cancellation, downgrade rate, support volume, and retention. For annual-default experiments, track cancellation before first renewal separately from ordinary monthly churn, since the economics and timing differ [4].
The Opportunity
3.53%
Illustrative new-MRR increase from a plan-mix shift alone, no change in conversion. In a purely illustrative model of 100,000 monthly pricing-page visitors at 3% conversion, shifting mix from 50/35/15 to 45/40/15 across a $20/$50/$100 ladder raises weighted ARPA from $42.50 to $44, worth $4,500 more MRR per acquisition cohort.
A Small Conversion Drop Can Erase the Entire Gain
If the same more-forceful recommendation reduces conversion from 3.0% to 2.9%, almost the entire apparent value of the plan-mix improvement disappears, this is why conversion and revenue must be read together, never plan-mix alone.
A "Winning" Test Can Still Be a Long-Term Loss
If customers steered into a highlighted plan cancel or downgrade faster because the package was a poor fit, the winning pricing test can eventually turn negative once retention is included [4].
Sample Size Alone Doesn't Guarantee a Trustworthy Result
GoodUI's Finn.com test had 55,293 visitors but only 5.3% statistical power at a 5% minimum detectable effect, while its Livefresh test with 272,793 visitors reached 88.8% power at the same threshold. A large-looking sample can still be underpowered [34].
Highlighting Changes Distribution, Not Traffic
A pricing highlight doesn't need new traffic, new product usage, or a new acquisition channel to change revenue, it alters the distribution of customers among packages that already exist, which makes it attractive but also easy to get wrong quietly.
Resources Worth Your Time