A rather unusual product designer

recruiter, 3D art school, marketer, UX/UI since 2023 after 8 years in product companies. Why? Because I love to understand people and drive business growth.My approach emphasizes continuous learning, clear communication, and tangible results.

A rather unusual product designer

recruiter, 3D art school, marketer, UX/UI since 2023 after 8 years in product companies. Why? Because I love to understand people and drive business growth.My approach emphasizes continuous learning, clear communication, and tangible results.

B2B BioTech | 0-1 Startup

Searching for an HIV Vaccine

Using product thinking to unify the tools building tomorrow's antibodies.

End-to-end Product Design
Acquisition
Retention
Web

To see work, let's chat >

B2C FinTech | Scaleup

Entering a World of Neobanks

Used positive friction to build trust and drive activation

Product Design
Activation
Mobile

See Case Study >

B2B2C AgeTech | Concept App

Concept Helping with Dementia

Built a UI Kit from scratch

Product Design
Mobile
Design System

See Case Study >

B2B ClimateTech | Startup

Bringing AI to Carbon Credits

Created website that raised seed funding and increase product demos

Product Marketing
Design
Desktop
Mobile

See Case Study >

B2B Open-Source | Corporate

Boosting Email Conversions

Researched across LATAM to uncover inclusive messages

Research
Content Strategy
Writing

See Case Study >

More Case Studies

As I meet different design teams, I thought I'd highlight other case studies I've genuinely enjoyed doing. Get in touch to see them.

Before becoming a product designer, I developed strong cross-functional skills as a brand marketer at BlaBlaCar 🚗 👥 facilitated workshops while teaching a Master’s Marketing lecture 📚 refined my product thinking while attempting to launch a tech product aimed at boosting Europe’s tech scene 🇪🇺 got good at interviewing being a recruiter for a few years in Singapore 🇸🇬 and built a small social enterprise close to my values 💜

🧃 The backstory

Trained in business, versed in different disciplines, I've slowly but patiently bled into the field of product design. What do I enjoy most? The superpower to show where we're going and wayfinding the path as a team.

🪣 Here are my skills bucketed

Product

Product thinking
Discovery
Data analysis
Ideating
UX design
Testing
Contributing to design systems

Marketing

Marketing strategy
Messaging
Copywriting
Creative concepts
Content producer
Campaign Manager

Talent

Employer branding
Account management
Cold outbound
Interviewing

General

Speaking to users is energizing
Enjoy one on one
Care for my team
Light up in workshops
Positive mindset
The storyline is critical
Care to learn and transmit

My mantra: the problem is what matters, tools are secondary

French and American, I’ve branched into digital product design after 8 years of experience across marketing, talent and product for mainly in-house tech companies - early stage to well established brands like BlaBlaCar and Automattic.Each project I've worked on had a powerful mission that has motivated me to live and work across 7 countries in three different languages. I like getting uncomfortable and having the chance to work with people that challenge the status quo - provided it makes business sense.Whether solving a new problem, preparing a storyline, running a workshop, or crafting prototypes on my laptop, I’ll always bring creativity, kindness, and focus on the goal.

🇺🇸 🇸🇬 🇬🇧 🇫🇷 🇨🇦 🇪🇸 🇪🇨 🇲🇽

Outside of work drawing, sculpting, photography and everything in between

Sports, family, friends, tandems

secret to my heart,
religieuses

Entering the World of Neobanks

An 8 day design challenge done with Lydia as part of the advanced training program with The Design Crew, Paris

Product Design
Mobile
8 days, Q1 2023

Lydia is a French mobile peer-to-peer payment app used by over 4 million people. The rise in neobanks has pushed it to get into banking.

Central to the change are joint accounts. The product team approached us to get recommendations to improve activation rates. This case study shows how I ran an end-to-end design process.

Felix Lepoutre, Lydia VP of Design

Results

  • Proved that positive friction and a progress system increased activation.

  • Reinforced trust through UX, UI, and writing.

  • Presented a recommendation deck to the VP and CPO of Lydia. They confirmed Lydia had taken a similar direction, "getting closer to the language and patterns of banks".

Process

Team

Discovery and Framing

From general problem, data and insights, to a problem statement

Brief in-take: by how much should activation rates go up?

Challenging a brief is essential. After evaluating the trade-offs, the target was revised to 30%.This was achieved by challenging the product team's tolerance with the radicality of the solution. The more change you hope for, the more the solution will likely be disruptive.

3 Activities

Research tools chosen based on the given constraints

1/ Interviews

Users revealed how Lydia has notoriety as a P2P payment solution, but not as a bank. Banks also have proven to have design patterns and language that differ from Lydia's identity.For example, they have much longer creation processes that are considered still "easy" and "fast".

See Lydia as a digital wallet and struggle to see it as a bank

3/4 participants

Consider other neo-bank onboardings fast and clear

3/4 participants

Deem joint accounts as a serious financial product compared to other Lydia services

3/4 participants

Goal

Motivations
Creation
Experience
Lydia Brand

Method

Screened Users
Semi-directed
4x30min, remote

Limits

Few users who've created a Lydia joint account

Link to Google Drive Research Plan, in french 🔗

2/ Audit

The creation flow of joint account is focused on acquisition, stopping short of having an activation logic.Also, out-of-app realities drastically slowed down activation. Marketing communication was wordy and unclear.

3/ Benchmark

It was noticeable that neobanks have key activation steps included in the creation process: payment method, top-up, and subscription.Their language was also clearly more banking.

Insights

Surfacing themes and insights from data collected

Trust

  • Traditional banks are trustworthy.

  • Lydia has brand debt, "Lydia’s a digital wallet at best"

  • Creating a joint account is a big step for couples and they guage provider's credibility.

  • Outside-app communication is fuzzy and hampers trust.

Motivations

  • Users creating a 'bank' account are quite motivated.

  • Delayed first purchases are in part due to "establishing rules of managing the account", "adding money", and other out-of-app realities i.e "delay in receiving payment card".

Organization

  • High mental load: information architecture and taxonomy are not consistent.

  • Confusion: interaction design is very low “Did I create the account?”, the value of joint accounts is unclear.

  • Pain: negative friction is present.

  • Objective-driven design: UX path stops abruptly upon creation.

Framing Tighter

Refining the initial problem to direct the design solutions

How might we create an actual banking experience that drives activation?

Success Criteria

  • 50% of users complete their first purchase in the first 7 days of creating the joint account.

  • 100% of users complete the onboarding within the first 7 days of creating the joint account.

We focused on the first use case: couples, because they are the main target user and to bracket complexity.

Ideating and Deciding

Searching for solutions that could answer our HMW

Workshops

After a Crazy 8, storyboarding, and mind-mapping, a set of ideas were created and plotted on an impact versus effort graph.

Go for a 'Guided Experience'

Central to the solution are UX writing language and design patterns common to banks.But also, to go beyond acquisition and define key steps that lead to activation. So to place them earlier in the creation flow.

1st Hypothesis

Adding friction to the creation process will be well received because couples take this seriously

2nd Hypothesis

People remember incomplete or interrupted tasks better than completed tasks, with the behavior of returning to the application to complete what is not finalized (Zeigarnik effect). Adding a progress system will accelerate activation.

3rd Hypothesis

Adding a budgeting tool can ease top-ups, a key activation step, because couples often struggle, out-of-app, knowing how much to add.

Guided Experience: Designing and Testing

Tweaking the creation flow and joint account UX

Tweaking the Creation Flow

Entering the banking territory, allows Lydia to stretch it's brand identity. Things can be slightly longer, in an effort to create trust through familiarity.

Pierre, Discovery Interview

Adding Positive Friction

In adding positive friction, by including in the creation process payment and top-up steps, it is hoped that users will feel in control and familiar with other bank patterns.This requires Lydia to stretch its brand identity from an experience that is always fast and easy to one that is tempered and thorough - like a bank!

Choosing payment method (left) and topping-up with budget widget (right)

Flowchart of creation process with two new steps, Link to Figma 🔗 (coming soon)

Other key changes

Yannick, Discovery Interview

Budgets to top-up

When adding money to a joint account, it’s common to ask ourselves “How much should I add?”. The tool aims to help users know how much they should add, while not having to remember, let alone calculate, what they decided out-of-app.

Florian, Discovery Interview

Clarify Writing

The prosaic intro page of joint accounts was changed by outlining the value proposition and using banking language to distinguish joint accounts from other in-app services.

Improving Joint Account Page UX

Once the account created, the user was confused as to what to do next and was also confronted to too much information.

Levahni, Audit

Adding a
Progress System

Regardless of the user path i.e. got a card, or not, topped-up, or not, the progress wheel tells users what encourages users with clear instructions that will eventually lead to a first purchase.

Homepage of joint accounts, a progress wheel is added to header

Information Architecture

To reduce mental overload, increase clarity, and drive trust, the welcome page got a tab navigation, common in other banking services.

Homepage of joint accounts; header is followed by navbar

Testing Plan and Results

Usability Test Overview

Questions

How will adding positive friction be perceived?

How does the progress system influence users upon creating the account?

How useful is the budgeting tool?

How does the navigation bar ease task completion?

Method

Semi-directed
4 x 30min, remote
Screened

Limits

Few users who've created an actual Lydia joint account

Link to Google Drive interview guide, in French 🔗

Learnings

Design Iteration

Improving the Progress System

Indicating progress was well received. The choice of a wheel could however be misinterpreted and block users.

Pierre, Usability Test

Sending the right signal

The progress wheel gave users the impression they couldn't use their account yet. It also forced a one-way next step, whereas cards provide flexibility to choose.

Homepage of joint accounts, instead of a progress wheel, cards are added to the header

Adrien, usability test

Cascading effects onto the visual identity

Lydia offers other services by presenting cards with stock imagery. This undermined the banking experience and we tweaked them to be more in line with the progress system.

Navy blue for progress cards infers authority and action from users.To create alignment and contrast, stock photos were removed and swapped for playful colors that suggested choice and exploration.

Second Prototype

Key hypotheses

1/ Positive friction will create trust and help Lydia's brand identity evolve towards a banking experience.2/ Adding progress cards accelerates activation. People remember incomplete or interrupted tasks better than completed tasks, with the behavior of returning to the application to complete what is not finalized (Zeigarnik effect).

Implementation

  • Phase 1, test new joint account creation experience: work with product marketing to drive joint account creations observe the change in activation rates.

  • Phase 2: drive adoption: ideate and test tools that add value and drive recurrent visits. For instance, to create a habit by providing value: a) capture recurrent spending, by asking to add a monthly bill b) suggest a budget to ease management of those bills and c) offer to optimize your budget by eliminating duplicates.

  • Phase 3: widen use cases: if the change is positive for couples, roll out similar solutions to secondary use cases. Research may be needed, or not.

Results

  • Proved that positive friction and a progress system would lead to an increase in activation.

  • Reinforced trust through UX, UI, and writing.

Felix Lepoutre, Lydia VP of Design

Wrapping Up

Learnings

  • Got more confident with UI, but still have a long ways to go, especially with font and color.

  • Simplicity is the underlying principle: let's not lose sight of the "why" of design, which is to comprehend the user, identify their issues, and then create a solution that meets those needs.

  • Prioritizing: have a clear first use case and stick to it to deliver faster rather than later. The rest is secondary.

  • On a personal note, after self-studying product design end of 2022, it was refreshing to work with and learn from talented mentors, my teammates Linh and Levanhi, while working on a real-world problem for one of France's top FinTechs.

Do Different

  • The solution for joint accounts I think also lies in the cross-functional work required to bring it to market correctly. Coordinating product and marketing would be an critical step.

Wrapping Up

Learnings

  • Got more confident with UI, but still have a long ways to go, especially with font and color.

  • Simplicity is the underlying principle: let's not lose sight of the "why" of design, which is to comprehend the user, identify their issues, and then create a solution that meets those needs.

  • Prioritizing: have a clear first use case and stick to it to deliver faster rather than later. The rest is secondary.

  • On a personal note, after self-studying product design end of 2022, it was refreshing to work with and learn from talented mentors, my teammates Linh and Levanhi, while working on a real-world problem for one of France's top FinTechs.

Do Different

  • The solution for joint accounts I think lies in the cross-functional work required to bring it to market correctly. Coordinating product and marketing would be an important step.

Thanks for reading.One more thing,
case studies are good conversation starters, agree?

If-so, Reach Out

See my work with BeMyCompass 🧭

Adapting cities to an aging population

A platform matching helpers and elders with Alzheimer's / elders reclaim autonomy / helpers strengthen communities / caregivers get support

Design System
UX/UI
Information Architecture
Prototyping

BeMyCompass grew from wanting to improve my UI skills.

Joining Ridd's cohort-based Figma training (Maven) brought me a wealth of knowledge and confidence. I got excited and explored a concept app addressing the problem my family is facing with my Mother's Alzheimer's.

Team

BeMyCompass grew from wanting to improve my UI skills.

Joining Ridd's cohort-based Figma training (Maven) brought me a wealth of knowledge and confidence. I got excited and explored a concept app addressing the problem my family is facing with my Mother's Alzheimer's.

Team

Dementia affects 55 million people worldwide.
By 2050, it's estimated to reach 150 million people.

How might we improve the lives of city elders with dementia, early to mid-stage?

Many plausible solutions.
This is one.

BeMyCompass is an intelligent navigation system allowing elders with dementia to walk their cities, freely and safely.

Discover BeMyCompass, through Isabelle’s story, my Mom.

It’s been 8 years Mom’s stuck at home. Healthy, but with Alzheimer’s, the city is too much to navigate on her own.

Today’s different, she has a walk.

Alone.

Her caregiver, my Dad, does a handover with BeMyCompass.

Mom starts her city walk while Dad gets on with his day, reassured.

BeMyCompass provides real-time tracking and keeps her within an intelligent area limit.

Mom explores, meanders, connects with the outside world.

She nears the area limit.

BeMyCompass notifies a network of helpers nearby. Trevor is the first to accept.

Finding her is easy. He nudges her away from the area limit.

BeMyCompass provides directions to helpers along her walk.

Helpers fasten ties with elders and are made aware of Alzheimer's.

BeMyCompass keeps track of each encounter.

Hours later, my Mom’s done her walk. Dad brings her back home.

Her spirit high, she rests on the couch.

BeMyCompass shares a day wrap with my Dad.

"Oh, you got to practice Arabic? How was that?"

She looks at Dad and smiles.

BeMyCompass is a concept app depicting cities where tech strengthens connection.

Process

Less talk, just show. Here are my Figma files. If we're both in it, just promise me you'll say hi 👋First time I build everything from scratch.

Foundations

Semantic colors

For a consistent look and feel across light and dark mode.

Product development

Driving design solutions from existing patterns through to information architecture upwards.

Learnings + Impact

There were many. Raising my comfort with visual design, but also thinking of the bigger picture.Product: how is this going to go market? What are the phases of development? What would be a simpler solution before becoming a marketplace? What are the different revenue lines?

UI/UX

Building a design system from scratch that isolates primitives from a style guide and product development file by using variables. Determining the happy paths and defining the components to build and integrate into the system.

Mental Models

Reinforcing my habit of identifying the one problem we’re solving at a screen level, finding patterns that already work, breaking them down, building my information architecture that incorporates these patterns, and then moving from low-fidelity up to high-fidelity design.I haven't highlighted testing, but prototyping when I wasn't sure helped me unblock, think the directions feature.

Myself

Getting more comfortable with craft and wanting to learn more. Building this presentation required some coding, which makes me think I'd be curious to learn more frontend.

Working on solving problems of an aging population? Want to iterate on this concept?

Get in touch

See my work with Kanop 🌳

Bringing AI to Carbon Credits

Product Marketing
Design
2022 & 2023

Kanop is an early-stage startup that uses AI to provide a software-as-a-service (SaaS) product for measuring, reporting, and verifying (MRV) carbon credits at scale.

The goal is to rebuild trust in the carbon market and maximize the potential of nature-based solutions, projected to support a third of the mitigation efforts by 2030. In 2022 and 2023, I wore a marketing and product designer hat to help Romain and Louis, co-founders of Kanop, go to market.

Romain Fau, Co-Founder Kanop

Results

  • I developed Kanop's early-stage brand and website, to bring a measuring, reporting, and verifying (MRV) product to the carbon credits market.

  • The journey was both challenging and enjoyable, sparking a desire to delve further into the ClimateTech space.

Process

Team

+30%

Creation of demo-accounts

Seed Funding

The work helped with messaging and therefore pitching to investors

x2.5

Average session time increase

Recruitment

Team grew from 2 to 10

Part 1

Discovery

Understanding the Problem Space

Visual and Written Language

I focused on identifying what words and concepts are used industry-wide in order to familiarize myself with the problem space.

Audience

  • Project developers who need MRV tools that are precise and affordable.

  • Forest managers who need to keep operational control for forests of all sizes around the world.

Benchmark

I looked at key messages, taglines, service offerings, and visuals. Unfortunately, testing products was difficult.What surfaced was some had a tone that was impersonal, therefore forgettable, while others raised potential conflicts of interest in the voluntary market.

Part 2

Minimum Viable Brand

Packaging brand essence, messaging and visuals

Brand Essence

The brand’s personality and tone of voice were developed. On the premise of breaking away from slow-dinosaur-like consultancies and embracing a modern tooling experience, it was decided to be somewhat earnest and fun while balancing a scientific approach.

Bobbing between remote workshops, the founders practiced articulating and crafting their vision, mission, and manifesto. Work that would soon enough bleed into all touchpoints whether in real life, the next pitch, recruitment, and many other digital touchpoints.

"Building a world where Earth's ecosystems are understood, protected, and integrated sustainably into tomorrow's economy and society."

- Extract from the Manifesto

Brand Messaging

Value Proposition

Kanop is focused on improving the measuring and reporting of carbon credits. Capabilities were paired with sub-problems and attributed a product feature.

Messaging Pillars

The underlying messaging and supporting selling points that precedes all content, whether it be copy or visual. A starting point that will need to be adapted as Kanop discovers its market.

Trust

  • Kanop only measures and report. No conflict of interest.

  • Scientists are the core of the team and are specialized in MRV and AI.

  • We're all about collaboration and data-sharing, so our API is accessible.

Ease

  • The product is designed to be like a SaaS. Night and day with consultancies.

  • Projects of all sizes are welcome.

  • MRV is fast. A matter of hours, not weeks.

  • Pricing removes the need for complicated budgeting.

Precision

  • In-house proprietary machine learning models are rigorously trained.

  • Customizable measurement levels. Prime focus on tree-level (suspected unique selling point)

  • Data sources are a mix of public and private.

Visuals

The last building block of the brand. Speed was of the essence, so colors, fonts, and logo were brought to a functional level.

A visual discovery phase focused on illustrating "tree-level", "trust, ease" and "precision" as they would be the pillars to the landing page.

Part 3

Landing Page

A first website with the goal to raise awareness

Architecture

To wireframe an efficient landing page, B2B landing pages of direct and, far-fetched, yet still very solid brands, like Apple Business, were analyzed.The overall structure led with a Why (vision), followed by Hows (Messaging Pillars), and What (supporting selling points). The whole, layered with trust e.g. social proof.For Kanop, this meant bringing tree-level measurement (USP), with precision, ease, and trust.

The Concept
"Meet Joe"

In an established market, focusing on product capabilities is key. However, many B2B SaaS brands have forgettable identities.To stand out, we designed a story-driven landing page with a hero "Joe", representing "tree-level measurement."

Personifying the tree emphasized Kanop's ability to bring users "up close and personal with billions of trees."

Other sections

Precision, ease and trust.

Testing

Gathering data from prospects, and investors, and looking at traffic gave insights as to what to tweak.The main takeaways:
a/ Kanop does not have a clear SaaS identity
b/ The aspirational messaging is too abstract, "Bye, Joe!"

The Good

  • People got curious with a raise in meetings and tickets.

  • Since launch, the need for trust in services that "only measure" is ever more important see Guardian article, 2023, TL;DR lots of carbon credits are being sold, yet they don't exist!

The Bad

  • The concept "Meet Joe" is not understood and requires digging. That's a red flag for a B2B SaaS entering a category.

  • Missing information relative to the product, terms of business, and screenshots are creating negative friction for conversion.

Questions

  • How unique is our selling point about providing tree-level measurement?

  • Is the experience consistent between our landing page and SaaS product?

How might we change the website for Kanop to be perceived as a SaaS?

Success Criteria

  • At least 20% increase in creation of demo-accounts.

  • Increase session times to over a minute.

Part 4

Reviewing Initial Design

Adapting architecture, using SaaS design patterns, and upping trust to drive product-led-growth

Architecture

The initial website aimed to raise awareness and interest, with a focus on tree-level measurement through aspirational messaging i.e. personifying a tree.With the feedback collected, a redesign of the underlying architecture was the starting point.Going from a single page to a use-case-driven architecture. Whereby the homepage would be broad enough to showcase both use cases and funnel visitors to their respective pages.

Reviewing the homepage

To dial up Kanop's "ease", SaaS design patterns were used. Untitled UI's design system was recommended for its flexibility and policy. Giving the team the ability to use similar patterns across the product and help create an experience that looks and feels whole.

By broadening the key message from "tree-level measurement" to "data analytics for nature-based solutions", the homepage now speaks to both audiences. Visitors are then funneled to their use-case page.

Highlighting Capabilities

As data and product development progressed, it was possible to offer essential details supporting the platform's features.

Deeper Trust

Using data to anticipate friction, an FAQ page was added with sections at crucial consideration points (home, features, use case pages, pricing).Testimonials from satisfied clients and team scientists reinforced reliability through social proof.

Other changes

  • Set up SEO foundation.

  • Provided Open Graph assets for social media sharing.

  • Fleshed out the team and pricing pages.

  • Oh, and breakpoints to make it responsive across devices!

Results and Learnings

The numbers and shared memories

A huge thank you to Romain, Louis, and the rest of the team - Antoine, Victor, Stéphanie (thanks for the monitor), Carla, and Nataly for their positive energy and collaboration!Reflecting on my journey, I first learned about carbon credits in 2013 from Alexander Yokoyama after a failed tech interview. The concept was abstract and even felt "wrong" as a pollution offset. However, over time, through involvement with La Fresque du Climat, Climate Collage and increasing climate change awareness, my understanding of the mechanism evolved. Today, I see it as an interesting solution by design.

+30%

Creation of demo-accounts

Seed Funding

The work helped with messaging and therefore pitching to investors

x2.5

Average session time increase

Recruitment

Team grew from 2 to 10

Wrapping Up

Learnings

  • Carbon credits, a solution that starts with good intentions, can easily go astray without proper controls. Technology can help maintain viability and address climate pressures.

  • The journey from ideation to Webflow was rich in learning.

  • Writing this case study taught me about product marketing, and I appreciate the insights from Anthony and Robert's work on early-stage B2B SaaS startups.

  • Balancing brand iteration and safeguarding is work since a brand is continuously evolving.

Do Different

  • Focus on one ideal customer profile (ICP) from the start: project developers.

  • Establish a clearer roadmap with founders to allow room for iteration and demonstrate the work's value.

  • Note to future self: refine marketing frameworks for workshops and deliverables to enhance clarity and efficiency.

  • Continue using Figma and improve the workflow from design to build with website builders e.g. Webflow.

  • Get better at visual development (font and color especially).

Wrapping Up

Learnings

  • Carbon credits, a solution that starts with good intentions, can easily go astray without proper controls. Technology can help maintain viability and address climate pressures.

  • The journey from ideation to Webflow was rich in learning.

  • Writing this case study taught me about product marketing, and I appreciate the insights from Anthony and Robert's work on early-stage B2B SaaS startups.

  • Balancing brand iteration and safeguarding is work since a brand is continuously evolving.

Do Different

  • Focus on one ideal customer profile (ICP) from the start: project developers.

  • Establish a clearer roadmap with founders to allow room for iteration and demonstrate the work's value.

  • Note to future self: refine marketing frameworks for workshops and deliverables to enhance clarity and efficiency.

  • Continue using Figma and improve the workflow from design to build with website builders e.g. Webflow.

  • Get better at visual development (font and color especially).

Facing brand and marketing challenges?

Would be happy to discuss

See my work with Automattic 📨

Research to Boost Email Conversions

I hope you'll find the hand drawings turn the dull into somewhat more of a story. I didn't want you to be bombarded with email screenshots.

Research
Content Strategy
Writing
Q2 2022

Every month, Automattic's open-source products are used by over a billion users. The team, 2,000 strong, distributed across 96 countries, recruits hundreds of engineers every year all over the world.

Central to the hiring process is attracting engineers with individual email outreach campaigns. However, at the beginning of 2022, the conversion rates for these campaigns significantly dropped.By incorporating a human-centered design approach to creating these campaigns, the trend was reversed.I took the lead in conducting research, fostering cross-team collaboration, and identifying conversion levers through testing.

Principal Consultant, Jason Brownlee

Results

2-3x Conversions

An uplift of 2 to 3x (year on year) across Latin America (LATAM). Other regions APAC and Africa showed similar early results.

DEI

Sourced insights from underrepresented groups (URGs) to test regional messaging and copy

Collaboration

Created new collaboration processes with the engineering team that benefited both recruiters, employer brand and hiring teams.

Scalability

The outline of these campaigns were easily replicable to other regions of the world.

Process

Team

Cross-functional collaboration with engineers, employer brand and other program leaders.

Part 1

Research

Primary Research | Secondary Research

Speaking to engineers

Interviews and a first workshop with engineers to critique outreach campaigns helped surface the following themes:a/ Depth: no added value. Everything can be googled.b/ Trust: a combination of things makes engineers wary of recruiters.c/ Writing: wording, tone, many elements created a sales feel that is particularly despized by engineers.

Regional realities

Secondary researched confirmed that personalization is key for campaign success. Interviewing underrepresented groups (URGs) within Automattic provided data and insights into regional nuances and their personal experiences at Automattic. The key themes:a/ The road to success
b/ A false sense of equity
c/ Often misled

Secondary Research

Previous campaign data showed:a/ A messaging focused on Automattic and used as a blanket template to all the regions of the world.b/ Tracking lacked structure to correlate changes with actual data.Also, features in the outreach tool, Gem, showed a swatch of opportunities for customization, but that's not necessarily creating trust and genuine personalization.

Part 2

Putting it Together

Insights and How Might We(s) | Solutions

Romina, Discovery Workshop

Writing style

The use of superlatives (e.g., "fantastic," "exceptional") contributes to a sales tone, and certain sentence structures come across as imposing. i.e "I don't want to bother you [...but I still will…] let me tell you why" can be off-putting.

How can we develop a writing style that is appreciated by our target populations?

Pablo, Engineer Interview

Trust

Recruiters are often met with skepticism in the job market due to unprofessional behaviors such as misrepresenting opportunities or acting as 'fromt the company' when in fact they are intermediaries.

How can we incorporate elements in our campaigns that enhance the credibility of recruiters?

Julian, Engineer Interview

Personalization

Emails currently provide minimal information about the engineer, relying mostly on details from LinkedIn. This approach limits what the recruiter can discuss, and generic hero content and job descriptions become too predictable.

How can we transform a luke-warm email into a warmer, more personalized one?

Senior recruiter, Interview

Tracking

As the talent team rapidly scaled, data tracking practices undermined reporting accuracy.

What would a basic tracking foundation look like?

How Might We Create a Trustable and Personalized Outreach in LATAM?

Success Criteria

  • Establish a conversion rate of a minimum 12% (previous year).

  • Campaign samples must reach at least 50 engineers in order to draw regional insights.

Solutions

Various solutions were explored across the team. With the given constraints around time and tooling, the focus was set on targeting and messaging.

Channel
Targeting
Messaging
CRM Parameters
Sender

Part 2

Crafting Better Campaigns

Starting with LATAM

Sharpen

To establish trust, strong arguments are essential. With the employer brand team, engineers, URGs, and program leaders a mapping of the value proposition with supporting talking points and content specific to LATAM, helped garnish the emails.

4 Variants
across 3 Audiences

The major problem is that when you speak to the world, there's an infinite amount of audiences.To start somewhere and, gradually, understand different engineering populations variants were tested.

(need to fix these gifs, work in progress, sorry!)

Chosen Audiences

a/ General Engineeringb/ Women Engineersc/ Service Agency Engineers

Variant Characteristics

a/ Adding more content.b/ Playing with length.c/ Having specific elements of value proposition for specific populations.

Tracking

To identify conversion levers and make sure the data was clean, a testing plan outlined control groups, variants, nomenclature and sample requirements.

The control group was a top performing standard campaign from 2021.

Results

Business Impact | Wrap-up

x2-3 Conversions

All variants in LATAM outperformed the control group with two reaching 3x improvements.

DEI

Except for women engineers, short emails were the most performing. Women engineers, however, were a lot more sensitive to long form.Targeting service agency engineers with our value proposition gave great results.

Co-designing

Automattic is asynchronous and collaboration can sometimes be difficult. Design tools brought engineers and talent closer as well as generating a lot of ideas for future projects.

Scalability

What was once siloed due to the growth of the team, iterating on messages and understanding the changes was made easy with variants. Rollouts to other regions, Asia Pacific (APAC) and Africa were easy and indicated similar findings.

Wrapping Up

Learnings

  • My favorite and proudest part of this project was going and speaking to our engineers who were not involved in the 'first contact.' Both engineers involved in hiring and those who are far from the process helped unearth insights and create a greater sense of belonging for both prospects and the team at large.

  • Operationally, I would have greatly benefited from the progress AI has brought, not necessarily to speed up the research, but rather to cover more ground and develop more compelling variants and iterations.

  • A more subtle but critical aspect was stakeholder management, which proved to be a lot more challenging than anticipated. I will take away from this experience that 'design thinking' requires education before people can fully adopt the principles.

  • Lastly, this approach had many caveats and areas for improvement. For example, when looking at 'regional realities,' it is easy to fall into stereotypes. To address this, greater involvement of underrepresented groups (URGs) in co-designing messages to hire more people like them could have been helpful!

Do Different

  • Priorities changed at Automattic and the iteration of the variants, as well as the rollout to different regions, slowed drastically.

Well-scoped research and design methods can be applied to any medium and add business value,
you agree?

Slot a Time to Discuss

See all case studies