← Back home

A shopping concierge experience

April to June 2017

Product Designer

Where it starts

Monsieur Harvey is a shopping concierge service helping tourists buy things and get delivered during their stay in Paris.

The idea came from our own experiences, families and friends coming from all over the world and sometimes need us to do a quick shopping for them.

Problem knowledge

Unfortunately, no one knows someone in every country and when they do, they don’t always get the help when they need the most.

After early research, we confirmed that during trips, those demands were specific and isolated in time. The context was key, they wouldn't ask us to buy souvenirs for example as it has a personal meaning.
On the opposite, when they don't find value in seeking things for hours in places they only know for a couple of days, they would rather rely on locals to buy goods they need.

France appears to be one of the best places to try to help people on this. In 2016, the Division of Major Enterprises (or DGE) said there was:

82,6 Millions


+40 Billions

Total incomes

Target audience

Large picture

We couldn’t address every tourist requests in every place in the world, we needed to choose a specific audience to validate our first product version and obviously learn from it.

When we review the tourism demographics to decide who will be that audience, the African population was one of the best to choose.

3,1 Millions

African tourists

13 nights

Average stay

2 Billions


We could actually touch a lot of people who were underserved on the market, those who were staying the longer during their trips. Meaning more potential needs so more opportunities for us.

Extract personas

We conducted a series of interviews of African tourists from differents countries, looking for similarities and differences, understand their motivations and expectations.

I took our findings and compile them into personas as below to help us empathize and build something they could rely on.

Mapping their trips

Going further in our knowledge of their context and needs, we frame the most common journey to evaluate which kind of experience would add value and where were the opportunities.

Making something too fancy would felt unnatural. They were pretty well organized as they came to stay quite long and didn’t want much but wanted their last resort if they needed something.

Then, we should lean on the simplicity felt by people asking their relatives or friends and bring high-end standard availability.

The concept

So far in our experiences, people would basically text us to ask because it's simple and inexpensive. To look easy, I wanted to stay close to what they were already doing and would accept: request something by texting and get delivered.

People will message us through a bot, Mr. Harvey, sometimes back by humans to keep the request as natural as possible. They will simply express what they want then shoppers will process it, do the shopping and deliver.

Every request will have a unique code that people will give to shoppers to validate the pickup and go.

Into the consumer app

v1 scope

We were concerned about making a product good enough to let us solve the problem and lead us to a product-market fit despite our lack of resources.

Considering what we could do and our first audience, three things appeared to be essential for the app:

Materializing our value proposition.

We needed to appear legitimate and trustworthy. One of our take on the app was to give invoices early in the app right after payment in addition to the one we give at the delivery time.

We thought we could use them to let crucial information available one swipe away during delivery and foster seamless interactions with shoppers.

The request's flow

The bot is the cornerstone of the request experience we are proposing.

There were few technical challenges before making it completely autonomous so based on our experiences, after few iterations on the script, I framed the most simple and natural request we could have.

The app will also subtly lead the conversation as a concierge would do to avoid failure and provide adapted controls to make response even easier.

User interface

Messaging apps having well-known patterns understood by people, the app volontarly appears like it is something they already know how to use.

With a limited set of colors, I wanted to convey simplicity but more importantly, a feeling of a premium high-end service. The interface should definitely be pleasant in use but evoke reliability and seriousness.

Also, in a constant effort to make it simple, I tried to optimize the way people should make their input by providing objects they could interact with and let them express their needs more efficiently.


A request we had for a perfume from private collection

The main screen is the messaging view. It’s our way to say there is nothing we could do better than help you with what you need, tell us!

Everything we wanted to express - ease, control, dynamism and proximity, should be felt the same way through the user interface, interactions and how the bot speak.

Supporting delivery experience

Users can see what they asked for and when we will deliver after the requested payment. The order code is also displayed so they don’t have to look out for the receipt into the app during pick up.

What now

While we are developing the app, we noticed from interviews on a larger target that if this service could help all those people, what could motivate them to use it differs regarding countries they were coming from.

So we pursuing researchs to understand all this, define how we talk to them and how we get traction even if they weren’t in our country

Word of mouth has been quite rewarding so far, we have already attract some early adopters that express request by simply texting us. We are excited to see if our work make their experience better.