Design Thinking

Buying Local During the Lockdown – A Service Design Study

From all the challenges we face with this quarantine, many initiatives have the good intention to help the small and medium business to survive. The site Le Panier Bleu is one of them. Join me in this exploration to review and reflect on what it takes to make the service last.

During the COVID-19 lockdown, a non-profit organization in partnership with Québec’s government launched a website: Le Panier Bleu. The website provides a listing of local companies to promote buying local and connect consumers to regional businesses.

Excited with the idea, my first experience visiting the site was frustrating. I could use the site to find a local business I already know, but Le Panier Bleu offered me no additional value or service over Google, for example. Using it to discover local companies made more sense to me, but the search offered by the site was not designed for product discovery. The whole experience was just too much effort.

Economic Solidarity

On the first weekend, the number of visitors from Québec looking for Le Panier Bleu crashed a homonym website in France, “Mon Panier Bleu.” Canadians already have a culture to look for and to consume local products as much as possible – the Made in Canada trend.

The vulnerability of small companies became very apparent in this crisis; the Made in Canada is now a solidarity movement to “buy local,” or to “buy from small business,” visible in other countries and all over social media.

Is it just me?

The new service got a lot of media coverage since its launch on April 5 2020. From the ten first results of a news search on Google, four articles can be classified to have a somewhat negative critique. A news article from Bloomberg mentioned the site was put together in two weeks, and there are plans to have an advanced version of the site soon.

There are many useful user insights, cultural specificities, and consumer behavior data to understand the need and the value of a service like Le Panier Bleu. To know a tale of the everyday user experience, read the story of a user trying to buy a dough cutter from a local business → Le Panier Bleu, j’essaie (in french only).

Going far

The site brings divided opinions. The development and funding constraints should have impacted the launch version, all amidst a sudden lockdown in Québec. Moreover, we have promises of new features soon.

I’m interested in exploring what would be a service that is not “too basic” or a service that can “go far” for the community.

This study is an ideation exercise applying design thinking techniques in a speculative scenario. The goal is to explore what would be a more useful but still feasible service. I do not have access to any internal information, resources, or technical constraints that might have affected the current site development.

The study

Using IBM’s Design Thinking framework, I started a 3 step process to:

  1. Observe: to understand the service, the cultural and economic background, and the end goal.
  2. Reflect: to answer “how Le Panier Bleu should be?”
  3. Make: to propose a concrete solution and recommendations.

Le Panier Bleu website is currently only in French; all references to items are freely translated to English for this project.

1. Observe

What is missing in the site Le Panier Bleu?

To understand what the site has, I broke down the statements in the ‘About’ page into exploration topics.

« Le premier objectif du Panier Bleu est de guider les consommateurs vers les commerçants québécois afin de stimuler leurs ventes. »

The goal is to drive sales to local stores. From this, we have two different target audiences – business owners and consumers in Québec.

How to bring consumers to business?

« …dans sa phase de lancement, se veut un répertoire complet des commerçants locaux. Un puissant moteur de recherche sera rapidement ajouté pour permettre de repérer les produits vendus chez nos commerçants… »

The initial stage of the website has the goal of showing a list of all businesses in Québec.

The key to connect consumers to business is the search. For the launch site version, this seems to be secondary. The search is only visible after the fold or if the user clicks the search icon.

On May 15, the site changed and placed the search visible first, yay!

How the search works now?

  • Search matches keywords in the business page. The business page does not have product listings, so there is no product search.
  • No search suggestion.
  • Instant results – no need to press, just start typing.
  • Filters available: category and location.

We can also see the total number of results found, a “reset search” button, a simple “previous/next” pagination, and only 10 results per page.

Search results algorithm always shows results per alphabetical order, no other sorting options (relevancy, proximity).

In the new site version of May 15, the search is the highlight of the interface. It shows a list side by side with a map of the closest stores to the visitor (based on IP location). The list of results has no pagination.

When using the search, the map expands the zoom level to show more results. Clicking on a circle in the map shows a business tooltip, but clicking the business card listing does nothing.

Another user’s story

Key takeaways from another user’s experience (in french):

  • The user reported being annoyed by the mismatched results (she tried to search for a “coupe-pâte,” a dough cutter, using multiple keywords and filter combinations).
  • Using the service reminded her of using yellow pages books.
  • The user impression is that there is good intention behind the service, but the site still needs to go much further.
  • There was too much effort to explore the site, changing keywords and location filters. Finding the results the user was looking for was not easy.

The business owner experience

Each business has a page with a summary, contact information, map location, social media, hours of operation, and delivery services.

  • The business owner needs to contact the site team to make any corrections or update their listing.
  • There is a waiting period for the form to be reviewed and appearing in the listing → 3 to 5 business days (comments on a Facebook post about this waiting time, with people reporting 3 weeks without response).
  • The business listing in the bottom has the text “Claimed Business” → It seems business can be claimed or unclaimed.

Technical stuff

The current website also made accessibility improvements, now the search and results can be fully accessible using the keyboard (it wasn’t in the first site version). Still, most buttons are missing state clues.

The website is capitalizing on the free media coverage, but it still needs to be found. Having an SEO strategy would only help the business listed. Some technical SEO and on-page SEO optimization notes:

  • No Robots, no sitemap, title tag, and H1 are not optimized.
  • It would benefit from having a page with a full list of business per region → improve local SEO
  • Business pages are now indexed in this version, yay!

Language: The site’s FAQ page mentioned they are working on an English version of the site.


The online search experience can take multiple shapes across users, but ultimately we start with awareness. When making a purchase online, we usually start with a broad search, then refine to a more specific search or maybe explore other options until we settle with a brand or store.

The current version of the website showing the stores on the map of my neighborhood is an excellent step towards recognition over recall. I know the streets and businesses around me, in general, should be easy to find these stores on Google Maps or other services. The real discovery is to connect the business to the consumer based on a purchase need.

By proving a list of businesses with categories and not products, Le Panier Bleu is increasing the user’s effort and now relying on recall. We need to make the connection from product to its wide-range category. For example: if I want to find a “chemise” (a shirt), I need to think about the relevant categories; should I search for “Vêtements pour hommes” (men’s clothes) or “Vêtements pour femmes” (women’s clothes). Again, not all searches are the same; some are made directly on categories; for example, when looking for a takeout meal, I would search first for a “restaurant.” A good search experience should comprise all scenarios.

For the business owners, there was a massive interest driven by media, which might have resulted in a sales increase. But with the website team as gatekeepers of information for any registry and page updates, it is hard to provide more features without increasing complexity and costs.

I would not expect to create personas to cover all behavioural search scenarios, for all use cases of the site. For this reason, I proceeded to create Jobs to Be Done (JTBD) statements based on anecdotal reports on social media and news articles. JTBD can uncover the same problems and needs for different groups so that I could focus on outcomes instead of features.

When I am craving for some specialty foods, 
I want to find local products and purchase them online,
so I can relive the experiences from my travels.

When I am at home during the quarantine
I want to continue buying from the same stores I usually do
so I can continue to enjoy the products I like and support the small business.

View more JTBD in this document.

Problem Statement


I’ll propose improvements as if following an agile approach, considering the features should be launched in stages (much like the current site is being updated now) and considering that personnel, technology, and infrastructure would impact the funds available.

During the Reflect stage, the goal is to brainstorm ideas, to ideate and sort the features into easy/difficult solutions vs. more/less value to users, and select a few to prototype during the Make stage. The suggested final product is a website delivering a relevant and unique service to the community.


This study intends to propose a long-lasting service, past the trending of economic solidarity of the current quarantine. To achieve this, we need to deliver a unique, pleasurable, and valuable experience to target groups (business and consumers) to keep the current trend of high engagement.

Strong Points

  1. Free media attention: the local government sponsors the website, thus having an innate news media coverage.
  2. High interest: “buy local” and “buy from small business” trends.

Pain Points

  1. Hard to discover specific services and products.
  2. Hard to update information (website admin gatekeepers).
  3. Differentiate service from current competitors:, search engines, maps services, Amazon.

2. Reflect

Based on the JTBD cards, we can surmise the most value for the user would be to have an e-commerce platform with multiple sellers and expedited delivery service. Le Panier Bleu is not a business site per se, a non-profit organization manages it, and intends to allow the local stores to increase profits.

Creating an e-commerce platform would require too many investments in all fronts and transform the site’s purpose. For this reason, I’ll propose changes without transforming the site into e-commerce.

Outcome #1 – save time for the user by offering advanced search capabilities:

  • Product and brand discovery;
  • Delivery service discovery;
  • Specialty product discovery (handmade, from local materials, organic, zero-waste, made in Québec, made in Canada);
  • Local vitrine with specialty products from the region;
  • Personalization, showcasing my preferred stores, my preferred products, and product suggestions.

Outcome #2 – increase sales:

  • Advanced business listing → add product, service, specialty markers, advertising within the platform;
  • Business environment → have a space with various resources to promote local stores into developing e-commerce business.

Feature Ideation

Offer a low effort search with advanced capabilities:

  • Search query with autocomplete and autosuggest.
  • Allow sorting → alphabetical, relevancy, proximity (add your postal code and find the results by proximity).
  • Numbered pagination.
  • Results view on map or list.
  • Filters → category, location, delivery, specialty products, type (business/product).
    • When displaying results filtering by business, no further action required to view delivery and contact information.
    • When displaying results filtering by product or brand, suggest stores closer to the visitor or e-commerce sites.
  • Zero-results page with listing suggestions.

Business-management related features:

  • Business zone → create an area behind a login screen, where business owners can update any information related to the store: products, services, locations, images, description.
  • Image gallery → create a database of reusable images.
  • Marketing solutions → a showcase of partners offering support services (e.g., design and advertising agencies, accounting services, HR services).
  • E-commerce solutions → a partnership with Canadian companies (e.g., Shopify, Canada Post) to offer e-commerce platform integration.
  • Training for business owners → offer business resources to stay afloat and make the e-commerce site thrive.
  • CSV / XML integration → manage products with .csv and .xml files.
  • Aliments of Québec → integration to fetch products in the external platform.

The Matrix

The solutions are evaluated being simple or complex, based on the questions:

  • An open-source technology offering the same capabilities exists?
  • The technology is ready out-of-the-box or requires minimum customization?
  • Does the solution need ongoing support from the website team?

The value to the user was estimated based on the Jobs-To-Be-Done cards: features mentioned in the cards have more value, while secondary or support features are assigned less value.

The search features are the ones that deliver more value to the user and are categorized as easy to implement. Without knowing the current environment and capabilities, we cannot know the complexity of these suggestions, so they are all guesses. Considering that the site already has a search engine implementation, I’m assuming the implementation considered scalability and standard search features (taking Lucene and Solr as reference).

The business-related features are more complex as they require new technology solutions, integration, and community building with other companies.

3. Make

The goal of this section is to explore the search features and business zone ideation. Creating a business zone in this stage would help to feed and enrich the search results, to expand to product search, and would remove the constraint of having a website gatekeeper to update business information. The website personnel would be focused now on support and compliance.

To re-make Le Panier Bleu with the new features, we also need a new wireframe:

With the basic structure defined, the prototype proposed is a desktop version, but mobile responsive. I’ve created only the main pages for the search functionality and the business zone. They are the ones with the most impact on the user experience.

In this proposal, the map view is not directly integrated with the search results, like the current version of the site. The user can choose to view results on a map. This goal of this approach is to shift the mindset of “local”: looking at the World Wide Web, we can consider all Québec my local area.

The business zone presents feature examples; however, the recommendation is to show to the users using progressive disclosure. We ask the user to fill out the basic listing first, and after the account setup, we suggest to add products, brands, delivery, and specialty tags.

I also chose to focus on searching instead or browsing through results. While browsing is something ubiquitous on social media sites/apps, search results help the user to achieve specific goals or answer specific questions. If the user wants to browse and use the service as a free discovery experience, they can still do so with filter controls.

Lastly, for this scenario/ideation to work, people are the key, especially the business owners community. The website administration shifts to a partnership; the store owners have more effort to update listings in the site, adding images and descriptions, but they also have more control and could update their listings themselves.

See the live prototype here!

And the single pages below.

Thank you for joining me on this journey, I had fun creating this study and I hope you enjoyed this exercise.

What would you do differently? For any comments please connect on LinkedIn or other social media channels 🙂