Skip to content
All work03/05 · Case study
Spicom · AI SaaS · CMS Marketplace

Watch the funnel.

Two pricing pages. One A/B test. Watch 27 visitors beat 12 in real time.

Spicom converted 12% of trial users to paid — well below benchmark. The old page was cluttered, plan differences were vague, trust signals were absent. The redesign moved the decision out of the pricing page and into the product. The simulator below replays the 90-day A/B test in three seconds.

Role
Product Designer · Research, design, rollout
Scope
Pricing & Billing · Conversion · Growth UX
Timeline
Mar 2023 – Jun 2023
Team
1 PM · 2 eng · 1 designer (me)

Auto-runs on first scroll · click Replay to run again

Live A/B simulator · 200 visitors · 90 days condensed
Visitor pool · 20050 / 50 split
OLD4 toggles · hard paywallcontrol
Card required to start
Paid · 12% conversion0 / 12
NEW3 plans · soft paywall + trustvariant
No card · 7-day trial
Trust signals · testimonials
Paid · 27% conversion0 / 27
A/B winner margin
OLD 0/100NEW 0/100+125% LIFT
Brief

Customers couldn't tell the plans apart.

Three billing toggles, vague feature descriptions, no money-back guarantee, no testimonials. Trial users were interested but unconvinced — they spent more time in the FAQ than reading the comparison.

Trial-user friction

I like the free plan, but I don't understand why I should upgrade. What exactly am I getting for my money?
Trial user · in-product survey

Funnel friction

55% of pricing-page visitors scrolled the page without engaging with any CTA — interested, but uncommitted.
Internal analytics · 90-day window
What was breaking

Three things blocked every conversion.

  1. 01

    Plan differences were invisible.

    Vague feature descriptions ('advanced support', 'premium features') told users nothing. 25% dropped off after viewing the comparison table.

  2. 02

    Trust signals were missing.

    No money-back guarantee, no testimonials, no transparent billing. B2B SaaS data shows 10–20% lift from visible trust signals — Spicom had zero.

  3. 03

    Five plans created decision fatigue.

    Successful competitors used 3–4 plans, side-by-side comparison, prominent trust signals. Five plans with overlapping features paralysed users.

Exhibit · the page Spicom shipped before

12% trial-to-paid

spicom.com/pricing

Pricing Plans

Choose the plan that's right for you

Free

$0

  • Feature one
  • Feature two
  • Feature three
  • Advanced support
  • Premium features

Starter

$9

  • Feature one
  • Feature two
  • Feature three
  • Advanced support
  • Premium features
Best

Premium

$19

  • Feature one
  • Feature two
  • Feature three
  • Advanced support
  • Premium features

Pro+

$49

  • Feature one
  • Feature two
  • Feature three
  • Advanced support
  • Premium features

Enterprise

Custom

  • Feature one
  • Feature two
  • Feature three
  • Advanced support
  • Premium features

Frequently Asked Questions

+ What's the difference between plans?
+ Can I cancel anytime?
+ Is there a refund?
+ Do you offer discounts?
+ What's included in advanced support?
⚠ no money-back guarantee · no testimonials · no plan-comparison

Three billing toggles. Five plans. Disclaimers louder than the offer. The pricing page was the funnel.

Three plans, not five

One axis. Three answers.

Three plans is the sweet spot. Five made users freeze; two felt insufficient. Three forces a clear hierarchy: entry · value · power.

  1. 01

    Simplify to 3 clear plans.

    Renamed to communicate purpose — Basic, Standard, Pro. Removed 3-month and 6-month toggles. Annual discount badge replaces a toggle.

  2. 02

    Lead with side-by-side comparison.

    Replaced bullet lists with a feature matrix. Hosting space became the primary axis — a single dimension users could read in one glance.

  3. 03

    Reframe Free as a value gate.

    Free got full theme installation and site testing. Only the custom domain was withheld. Upgrade trigger is launch — not deprivation.

  4. 04

    Anchor every change to an A/B test.

    Money-back guarantee, testimonials, transparent billing — each had a research insight behind it and a measurable hypothesis ahead.

What we tested · what moved conversion

Every change had a hypothesis and a measurable outcome.

Recruiter intuition would have asked for bigger headlines. The data asked for testimonials, money-back guarantees, and a premium trial. Each lifted conversion measurably; the table below is the full ledger.

  1. Hypothesis

    Trust signals near the CTA increase conversion.

    Variant · Added testimonials + money-back guarantee inside the Pro card. Removed FAQ section above pricing.

    +18%trial-to-paid
  2. Hypothesis

    Soft paywall outperforms hard paywall.

    Variant · Removed the 'card to start' requirement on Free signup. Card requested at upgrade intent only.

    +47%trial-to-paid
  3. Hypothesis

    Three plans beat five.

    Variant · Collapsed Basic / Standard / Pro / Pro+ / Custom into Free / Pro / Business. Realigned to capability axis.

    +21%conversion · +29% comprehension
  4. Hypothesis

    Premium trial reduces post-conversion churn.

    Variant · Pro plan auto-includes a 7-day premium trial. Email reminder day 6.

    −18%churn · +20% MRR

14

Variants tested

4

Variants shipped

2.25×

A/B winner margin

The exact upgrade moment we redesigned

Six stages. One decision.

The redesign moved the decision out of the pricing page and into the product. Step through the moment as a merchant would — limit, soft paywall, plan compare, trust, card form, welcome.

Spicom · Dashboard · Free plan

Stores

store-auroraActive
Stores used1 / 1

You've reached your store limit.

Free plan includes 1 store. Add another store to scale. See plans

Stage 01 · Design decision

The trigger lives in the user's flow.

Limits appear inside the dashboard the user already opened — not on a marketing page they have to navigate to. The upgrade story starts where the work happens.

What changed after launch

Trial-to-paid more than doubled.

Tracked over the first 90 days post-launch through analytics, in-product surveys, and revenue dashboards.

+0%

Hero metric

Trial-to-paid conversion lift

12% → 27% over the 90-day post-launch window. Same acquisition spend, same traffic. The page changed; the funnel converted.

12% → 27%Industry benchmark · 18%Spicom internal benchmark since
  • +35%

    Time on pricing page

    users were thinking

  • +40%

    Plan-selection CTA clicks

  • −18%

    Churn rate

  • +20%

    Monthly recurring revenue

  • 2.25×

    A/B test margin

    redesign vs control

Stakeholder · 90-day retro

“This redesign had an immediate impact — we've never seen this level of conversion before. It's now our benchmark for future product changes.”
— Product Manager, Spicom
Reflection

Four lessons that traveled with me.

  1. 01

    Trust signals near the CTA outperformed every visual hierarchy change we tested. Recruiter intuition would have asked for bigger headlines; data asked for testimonials.

  2. 02

    The 7-day premium trial reduced both pre-conversion hesitation AND post-conversion churn. Letting users feel the value before buying is the most under-used SaaS pattern.

  3. 03

    Three plans is the sweet spot. Five made users freeze; two felt insufficient. Three forces a clear hierarchy: entry · value · power.

  4. 04

    Tracking time-on-page alongside conversion is essential. Time went up AND conversion went up — proving users were thinking, not bouncing.

All workEnd of case study · 03/05