eMazing-retailing

- Introduced innovative market research methods focused on A/B testing for FMCG products.
- Led a team of researchers and analysts to deliver data-driven insights for better decision-making.
- Helped clients optimize product offerings and marketing strategies based on direct consumer feedback.
- Developed proprietary tools and methodologies to enhance market research accuracy.

Co-Founder eMazing-Retailing

Procter & Gamble

- Unveiled consumer insights and optimized market positioning for new Lacoste fragrance through A/B testing and data mining.
- Deployed a tracking and response process for a newly launched Stella McCartney fragrance, converting crowdsourced insights into quantitative data.
- Identified key assets to improve in-store and e-commerce performance using consumer research.

Consumer Market Knowledge Manager Fine Fragrance PROCTER & GAMBLE

- Supported global restaging of Ariel/Tide, revealing a $100 million opportunity in Europe.
- Conducted market and consumer research in emerging markets to strengthen Tide’s brand positioning.
- Led strategic thinking and market research in China for new detergent concepts.

Brand- Manager Consumer Market Knowledge Global Fabric Care

- Identified top opportunities in the Western Europe Fabric Care market and guided project priorities.
- Deployed a specific process to track and optimize detergent expansion in Germany and other countries.
- Led digital marketing mix modeling projects to drive non-TV business models.

Brand - Associate Consumer & Market Knowledge Western Europe Fabric Care

Lidl

- Supervised three stores and managed a team of store directors.
- Implemented strategies to improve store performance and customer satisfaction.

Retail Store Manager

Schneider Electric

-Analyzed the financial situation of key suppliers to prevent potential supply disruptions.

Consultant

Grenoble Ecole de Management

- Managed strategic projects for companies in the Rhone-Alpes region.
- Developed a strategic plan to boost sales for KING-JOUET stores, receiving top recommendations from the CEO.
- Audited a company in recovery proceedings and set up a business plan for a French company expanding into Romania.

Junior Project Manager (internship), MASTER IN ENTREPRENEURSHIP

Katalyse

- Conducted market, strategy, and financial analysis for various companies.
- Presented research findings to clients.

Junior Consultant (internship)

Junior ESC Grenoble

- Managed a client portfolio and conducted market research for companies like La Poste and Grenoble Airport.
- Recruited, trained, and managed teams for market survey projects.

Sector Research Manager

We help large companies like Procter & Gamble, L’Oréal, and Logitech… do better more predictable market research using the principles of A/B testing

Want to know how top brands stay ahead in market trends?

In 2017, I started Emazing Retailing to change how companies approach market research. 

With a deep interest in understanding consumer behavior, 

→ I wanted to use data to help businesses make better decisions. 

We use A/B testing in market research to get clear and direct feedback from consumers.

I lead a talented team dedicated to helping businesses succeed. 

Our goal is to give companies the tools and insights they need to understand the market and make better decisions.

Here’s how we helped our clients:

→ Improved market positioning for a new Lacoste fragrance using A/B testing and data analysis.

→ Set up a successful tracking and response process for Stella McCartney fragrance.

→ Found a $100 million opportunity for Ariel/Tide in Europe.

→ Introduced a new way to test eCommerce assets for FMCG products at Emazing Retailing.

***

Learn more about how we can improve your market research at [emazingretailing.com](http://emazingretailing.com/).

Or book a 1:1 call with me (click on my featured section )

📞 Let's chat: https://calendly.com/peberthier-1/emazing-discussion

Standard A/B testing doesn’t tell the full story of what makes a product win on Amazon.

Here's what actually determines product success:

1) Competition matters

Your product doesn’t live in a bubble. Buyers are comparing it against dozens of other options.

Success depends on how you perform in that context, not just on isolated test numbers.

2) Shopper behavior counts

78% of customers look at multiple products before buying.

The first 3 seconds decide if they click. Your images, titles, and overall look drive those choices.

Here’s what really shows performance:

The metrics that matter most:

• Click-through rates vs. category averages
• Conversion rates when shown beside competitors
• Shopper paths during the buying process

Top brands don’t just test alone.

They test in real-world settings, where customers have alternatives and must make choices.

This shows insights that normal A/B tests miss:

→ How you stack up against competitors.
→ How buyers perceive your spot in the category.
→ Why they pick you, or not.

At eMazing Retailing, we help brands test in these competitive settings, track shopper actions, and improve based on real market data.

***
Want to know how your products truly stand up to the competition?

Book a free demo (link in comments) and see what might be missing in your Amazon strategy.

Pierre-Emmanuel Berthier

We help large companies like Procter & Gamble, L’Oréal, and Logitech… do better and more predictable market research using the principles of A/B testing

Many companies use virtual shelves to test new packaging, products, or ideas.

But this method has big flaws that can lead to misleading results.

Here are four main problems:

1. Limited device reach

Testing only on computers or tablets leaves out the many people who mainly use phones to shop.

2. Demographic bias

People who don’t use computers often get excluded, so your sample isn’t a true reflection of your market.

3. Fake shopping experience

Clicking through a digital “shelf” on a computer doesn’t feel like real shopping and doesn’t match natural decision-making.

4. Higher research costs

Finding participants with the right devices is more costly and shrinks your sample size.

But testing in e-commerce settings works better:

→ People already shop online, so it feels natural
→ Any device can be used, making it more inclusive
→ The process mirrors real buying behavior
→ Results cover brand, packaging, and price choices in one go

At eMazing-retailing, companies have seen 40% more accurate results with e-commerce testing compared to virtual shelves.

***
Are you confident your current testing reflects real behavior?

E-commerce methods may give you a clearer view of how products will actually perform.

If you need any help, book a consultation with us: https://lnkd.in/eppXiSSi

Behavioral data doesn't mean identifying people by name or collecting anything that could reveal their personal identity.

Instead, it's about tracking, second by second, what appears on their screen, what actions they take, and ultimately what they purchase.

The information we focus on is typically contextual and demographic:
whether they're using a computer or a mobile phone, their age group, the city they live in, and other sociodemographic factors → not personal identifiers.

Our interest isn't in individuals as isolated cases, but in cohorts… groups of people who share similar profiles, such as age range, location, brand preferences, or shopping habits.

The value lies in identifying trends and patterns that emerge across dozens or hundreds of shoppers, not in pinpointing any single person.

At the end of the day, it's the shared behaviors and significant shifts in group behavior that matter(not the identity of any one individual).

This methodology delivers powerful insights about consumer behavior while maintaining privacy.

We learn how different demographic groups:

→ Interact with products.
→ What drives their decisions.
→ How to optimize experiences for broad segments.

The result is actionable business intelligence without compromising individual privacy.

***
How are you balancing the need for consumer insights with privacy protection?

The right approach can deliver powerful behavioral data while respecting individual privacy boundaries.

One of the most surprising tests we conducted was for a well-known fast-food chain.

We recreated the interface of a typical ordering kiosk, similar to the ones you now see in most fast-food restaurants.

The goal was to determine the most profitable menu change to increase sales.

The change we tested was simple:

At the end of the ordering process, we suggested additional products for customers to add to their order.

While many customers did choose the suggested extras, a comparison with a control group (without the suggestion) revealed something counterintuitive:

Customers who accepted the upsell ended up compensating by spending less on other items they would have normally chosen themselves.

Even when we successfully nudged people toward the products we wanted them to buy, they balanced their spending by cutting back elsewhere.

This was fascinating for two reasons.

First, it showed how realistic our experiments are:

Even though participants were not ordering with real money, they behaved exactly as if they were → carefully managing their "spending."

Second, it revealed just how aware customers are of their overall spend.

If we got them to spend more on one item, they instinctively adjusted their 
choices to avoid increasing their total bill.

To validate the results, the client ran the same menu change in a pilot restaurant.

The same compensating behavior appeared in the real-world setting, confirming both the accuracy of our test and the realism of our experimental approach.

***
P.S.:If you're curious about how realistic behavioral testing can reveal unexpected consumer patterns, our methodology might uncover insights that traditional research would miss.

Book a consultation with us: https://lnkd.in/eppXiSSi

Most brands test prices the same way they test packaging or concepts( by showing each option separately).

This method, called monadic testing, gives clearer insights than standard tools like Gabor-Granger or PSM.

Traditional pricing models often produce neat bell-shaped curves that look nice but don’t reflect reality.

They assume interest steadily rises as prices fall, then drops again.

In practice, interest can spike at different price points because of how competitors are priced.

Here’s the monadic approach:

The only way to see these spikes is by testing each price point on its own.

Clients bring several price ideas, and we test each separately.

If they don’t have those, we run a two-step behavioral test:

1) Build a simulated store where every product shows a price (except the one being tested).

2) Ask participants to set their own price and then decide if they’d buy at that price.

By keeping pricing tests monadic and behavior-based, we get results that are closer to real buying behavior.

***
Are your pricing decisions based on smooth charts that look good, or on real-world market responses?

One of the biggest blind spots in Amazon selling is relying too much on A/B testing.

Since Amazon makes it easy, many brands focus on tweaking small things(like one product image) without seeing the bigger impact.

On Amazon, you can’t run linked A/B tests across multiple products.

A change that helps one SKU might actually pull sales away from another.

For example, improving one listing could shift attention and reduce sales for products in the same line.

But here is a better approach:

To understand the effect across your whole range, you need to test outside of Amazon.

That way, you can control the setup and see how changes impact your entire portfolio, not just one product.

Here’s why this matters:

Testing only individual products can lead to short-term wins but long-term loss.

Growth comes when you optimize the portfolio as a whole, not when products are competing against each other.

True optimization means looking at how your products perform together.

It’s about making sure changes help your full range (not just one listing at the cost of another).

***
Are you testing single products or your entire portfolio?

The answer decides whether you grow sustainably or stall by accident.

AI predicts attention, but can't predict sales.

Here’s the gap that’s hurting brands:

We all know that AI is changing fast.

What works today might not work tomorrow.

One area it’s strong in right now is predicting which parts of a package grab attention.

Trained on huge sets of eye-tracking heatmaps, AI can show where people are most likely to look.

But here’s the catch:

Attention doesn’t always mean interest, or sales.

In tests, some packages got high attention for the wrong reasons. They stood out but weren’t the ones people bought most often.

That’s why AI works well as a tool to guide decisions, but it shouldn’t decide on its own which designs go to market.

AI is especially useful in speeding up early creative work.

It can quickly turn rough text ideas into draft designs or product images.

These can then be tested in research or used online.

AI is a strong support for market research.

It makes the process faster and richer, but it doesn’t replace human judgment or real shopping data.

***
So how are you using AI? As a support to make better calls, or as the only voice in the room?

Here’s why half the market research industry might be left behind

Different companies treat market research in different ways.

Some are already using better methods that look at how people actually behave.

Others are still picking research suppliers based only on price and how much data they get.

Right now, two big trends are shaping the market research space:

1) Fast and cheap DIY tools:

These tools use old-school methods, but at a larger scale and lower cost.

They give companies lots of data quickly, and often come with features to 
help sort through all that information.

2) Focus on better data, not just more:

Many suppliers now understand that collecting data is not enough.

What really matters is whether that data reflects real-life behavior.

There’s a big difference between what people say in a survey and what they actually do in a store.

3) Soon, we might see a third trend:

Tools that are fast and affordable like DIY platforms, but also give deep, real-world insights like behavioral research does.

So here's the question:

Is your company just collecting more and more data?

Or are you focused on the kind of data that can actually predict what people will do?

***
If you're curious about how behavioral research might help you, we have a simple, effective method that blends real behavior with easy-to-use tools.

Book a consultation with us to know more: https://lnkd.in/eppXiSSi

Taplio

Pierre-Emmanuel Berthier's Linkedin Analytics

's Best Posts (last 30 days)

Want to drive more opportunities from LinkedIn?

Taplio

.css-1jgreyo{background-image:linear-gradient(284deg, rgba(39,43,72,1) 0%, rgba(33,47,183,1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Pierre-Emmanuel Berthier's Linkedin Analytics

's Best Posts .css-qqfgvy{font-size:var(--chakra-fontSizes-md);}(last 30 days)

Want to drive more opportunities from LinkedIn?

Pierre-Emmanuel Berthier's Linkedin Analytics

's Best Posts (last 30 days)