Download CV

Boosting Pricing page conversion through UI/UX Optimization

Client:

NDA

My role:

Product designer

Team:

Product designer, BA, Product Manager

Duration:

3 month

Design impact

Book‑a‑call rate

Before

 3.2 %

After

6.8 %

Impact

110%

Form‑completion rate

Before

28 %

After

52 %

Impact

85 %

Pricing‑page bounce

Before

62 %

After

38 %

Impact

24 %

Form‑finish time

Before

4 m 20 s

After

1 m 50 s

Impact

58 %

Before

After

Problem

  • Sales reported that 70% of leads were unclear on how the service integrated with their current processes.
  • Marketing highlighted a 62% bounce rate on the pricing page.
  • Management set an ambitious target: boost the conversion rate to 10% within three months.

User Behavior Analysis

  • Heatmaps (via Hotjar) showed that only 30% of users reached the case studies section.
  • Session recordings (via Microsoft Clarity) revealed that 40% of users clicked on non-interactive elements, signaling unclear UI cues.
  • On the pricing page, 50% of visitors left after viewing the pricing tiers without engaging with the “Book a call” CTA.
  • Form analytics indicated that 60% of users abandoned the call booking form at the “Tell us about your company” field due to information overload.

Solution

To fix that, I built and demoed three radically different pricing concepts:

  • Option 1 – Configurator gridVisitors pick a single outbound channel, adjust volume with steppers, and see the price update live.

 

  • Option 2 – Full custom bundleA sidebar lets buyers combine any outreach channels they want; the quote recalculates on the fly.

 

  • Option 3 – Two clear, static packages ★Plain comparison table: “Omnichannel” vs “Omnichannel Plus”, each framed by the yearly appointment volume.

Option 1

Option 2

Option 3

Results

Options 1 and 2 were slick and interactive, but five quick hallway tests exposed a problem: prospects spent more time fiddling with controls than understanding value. The simplest design - Option 3 - won every clarity checkpoint (“I get it immediately” in 4 / 5 sessions) and needed almost no engineering effort.

Engineering spent just 14 hours swapping the old grid for the new two‑plan layout, neutralising the palette, and adding a sticky bar.

 

8 weeks post‑launch, booked calls had more than 2x, bounce rate fell by 3x, and the booking form took less than 3 minutes to finish.

Take away

When users are drowning in choices, reducing interactivity can be the most powerful UX decision you make.

Next project

Let's create something

awesome

together

Boosting Pricing page conversion through UI/UX Optimization

Client:

NDA

My role:

Product designer

Team:

Product designer, BA, Product Manager

Duration:

3 month

Design impact

Metric

Before

After

Impact

Book‑a‑call rate

 3.2 %

6.8 %

110 %

Form‑completion rate

28 %

52 %

85 %

Pricing‑page bounce

62 %

38 %

38 %

Form‑finish time

4 m 20 s

1 m 50 s

58 %

Before

After

Problem

  • Sales reported that 70% of leads were unclear on how the service integrated with their current processes.
  • Marketing highlighted a 62% bounce rate on the pricing page.
  • Management set an ambitious target: boost the conversion rate to 10% within three months.

User Behavior Analysis

  • Heatmaps (via Hotjar) showed that only 30% of users reached the case studies section.
  • Session recordings (via Microsoft Clarity) revealed that 40% of users clicked on non-interactive elements, signaling unclear UI cues.
  • On the pricing page, 50% of visitors left after viewing the pricing tiers without engaging with the “Book a call” CTA.
  • Form analytics indicated that 60% of users abandoned the call booking form at the “Tell us about your company” field due to information overload.

Solution

To fix that, I built and demoed three radically different pricing concepts:

  • Option 1 – Configurator gridVisitors pick a single outbound channel, adjust volume with steppers, and see the price update live.

 

  • Option 2 – Full custom bundleA sidebar lets buyers combine any outreach channels they want; the quote recalculates on the fly.

 

  • Option 3 – Two clear, static packages ★Plain comparison table: “Omnichannel” vs “Omnichannel Plus”, each framed by the yearly appointment volume.

Option 1

Option 2

Option 3

Results

Options 1 and 2 were slick and interactive, but five quick hallway tests exposed a problem: prospects spent more time fiddling with controls than understanding value. The simplest design - Option 3 - won every clarity checkpoint (“I get it immediately” in 4 / 5 sessions) and needed almost no engineering effort.

Engineering spent just 14 hours swapping the old grid for the new two‑plan layout, neutralising the palette, and adding a sticky bar.

 

8 weeks post‑launch, booked calls had more than 2x, bounce rate fell by 3x, and the booking form took less than 3 minutes to finish.

Take away

When users are drowning in choices, reducing interactivity can be the most powerful UX decision you make.

Boosting Pricing page conversion through UI/UX Optimization

Client:

NDA

My role:

Product designer

Team:

Product designer, BA, Product Manager

Duration:

3 month

Design impact

Metric

Before

After

Impact

Book‑a‑call rate

 3.2 %

6.8 %

110 %

Form‑completion rate

28 %

52 %

85 %

Pricing‑page bounce

62 %

38 %

38 %

Form‑finish time

4 m 20 s

1 m 50 s

58 %

Before

After

Problem

  • Sales reported that 70% of leads were unclear on how the service integrated with their current processes.
  • Marketing highlighted a 62% bounce rate on the pricing page.
  • Management set an ambitious target: boost the conversion rate to 10% within three months.

User Behavior Analysis

  • Heatmaps (via Hotjar) showed that only 30% of users reached the case studies section.
  • Session recordings (via Microsoft Clarity) revealed that 40% of users clicked on non-interactive elements, signaling unclear UI cues.
  • On the pricing page, 50% of visitors left after viewing the pricing tiers without engaging with the “Book a call” CTA.
  • Form analytics indicated that 60% of users abandoned the call booking form at the “Tell us about your company” field due to information overload.

Solution

To fix that, I built and demoed three radically different pricing concepts:

  • Option 1 – Configurator gridVisitors pick a single outbound channel, adjust volume with steppers, and see the price update live.

 

  • Option 2 – Full custom bundleA sidebar lets buyers combine any outreach channels they want; the quote recalculates on the fly.

 

  • Option 3 – Two clear, static packages ★Plain comparison table: “Omnichannel” vs “Omnichannel Plus”, each framed by the yearly appointment volume.

Option 1

Option 2

Option 3

Results

Options 1 and 2 were slick and interactive, but five quick hallway tests exposed a problem: prospects spent more time fiddling with controls than understanding value. The simplest design - Option 3 - won every clarity checkpoint (“I get it immediately” in 4 / 5 sessions) and needed almost no engineering effort.

Engineering spent just 14 hours swapping the old grid for the new two‑plan layout, neutralising the palette, and adding a sticky bar.

 

8 weeks post‑launch, booked calls had more than 2x, bounce rate fell by 3x, and the booking form took less than 3 minutes to finish.

Take away

When users are drowning in choices, reducing interactivity can be the most powerful UX decision you make.

Next project

Let's create something

awesome

together

Boosting Pricing page conversion through UI/UX Optimization

Client:

NDA

My role:

Product designer

Team:

Product designer, BA, Product Manager

Duration:

3 month

Design impact

Metric

Before

After

Impact

Book‑a‑call rate

 3.2 %

6.8 %

110 %

Form‑completion rate

28 %

52 %

85 %

Pricing‑page bounce

62 %

38 %

38 %

Form‑finish time

4 m 20 s

1 m 50 s

58 %

Before

After

Problem

  • Sales reported that 70% of leads were unclear on how the service integrated with their current processes.
  • Marketing highlighted a 62% bounce rate on the pricing page.
  • Management set an ambitious target: boost the conversion rate to 10% within three months.

User Behavior Analysis

  • Heatmaps (via Hotjar) showed that only 30% of users reached the case studies section.
  • Session recordings (via Microsoft Clarity) revealed that 40% of users clicked on non-interactive elements, signaling unclear UI cues.
  • On the pricing page, 50% of visitors left after viewing the pricing tiers without engaging with the “Book a call” CTA.
  • Form analytics indicated that 60% of users abandoned the call booking form at the “Tell us about your company” field due to information overload.

Solution

To fix that, I built and demoed three radically different pricing concepts:

  • Option 1 – Configurator gridVisitors pick a single outbound channel, adjust volume with steppers, and see the price update live.

 

  • Option 2 – Full custom bundleA sidebar lets buyers combine any outreach channels they want; the quote recalculates on the fly.

 

  • Option 3 – Two clear, static packages ★Plain comparison table: “Omnichannel” vs “Omnichannel Plus”, each framed by the yearly appointment volume.

Option 1

Option 2

Option 3

Results

Options 1 and 2 were slick and interactive, but five quick hallway tests exposed a problem: prospects spent more time fiddling with controls than understanding value. The simplest design - Option 3 - won every clarity checkpoint (“I get it immediately” in 4 / 5 sessions) and needed almost no engineering effort.

Engineering spent just 14 hours swapping the old grid for the new two‑plan layout, neutralising the palette, and adding a sticky bar.

 

8 weeks post‑launch, booked calls had more than 2x, bounce rate fell by 3x, and the booking form took less than 3 minutes to finish.

Take away

When users are drowning in choices, reducing interactivity can be the most powerful UX decision you make.

Next project

Let's create something

awesome

together