This concept isn't new. Big shots like Google, Microsoft, and Meta have thrown it around before. At FARFETCH, though, it's not just some buzzwords; it's rooted in our commitment to "Amaze Customers." Recognizing that not all team members are enthusiastic luxury shoppers, we see this as a distinctive chance for them to immerse themselves directly in the customer journey.

This hands-on experience, coupled with existing methods like user research, site visits, and engaging conversations with boutiques and customers, enriches their understanding of the journey they actively help mould. This philosophy underpins the inception of our dedicated dogfooding program.

The dogfooding philosophy

So, what exactly is dogfooding? Let's set aside the literal interpretation. It's the practice of immersing oneself in the experience of a product or service. Interestingly, it originated from an invitation to dog owners to try the food they were feeding their dogs, showcasing the confidence Alpo had in its product back in 1976.

For a product-led company, dogfooding is the act of team members personally testing the digital products they offer to customers. It involves experiencing the product's quality, anticipating potential issues, and exploring pitfalls.

To be clear, it doesn't entail consuming non-human vetted food; rather, it's about stepping into the customer's shoes to understand their perspective. The beauty of it all? Your team or company can seamlessly implement this approach right away, requiring no additional budget.

During this process, the focus is on identifying bugs and gathering suggestions for improvement. Establishing a systematic approach is key to promoting a culture that encourages the discovery, reporting, and prioritization of issues. The ultimate goal is to foster a product culture oriented to real-world usage of our products, elevating our customer experience.

To dogfood

It’s important to clarify what’s expected from such a program. Dogfooding is:

  • A hub for quality culture, raising the bar for the luxury experience our customers expect.
  • A space to voice feedback on hiccups and offer suggestions for smoother customer journeys.
  • Not a panic room for high-priority meltdowns (like payment methods going rogue).
  • It does not replace our beloved QA teams’ work

For us, this program is anchored in three main concerns:

  1. Build trust: We have an estimation of daily 3.5M visitors to our website daily, with a customer experience that we must be obsessed about.
  2. Foster a culture of quality: Imagine a building with broken windows – a metaphor for visible disorder in a neglected environment. If one window remains unrepaired, the others will follow suit.
  3. Improve usability: We picked up on our customer' knowledge and listened to their feedback. Our team is mostly focused on tech, so it may not be biased and fully represent our diverse customers.

Implementing new processes in a busy product team? Tough stuff. I hear you thinking, "So my overloaded team will magically adopt this?" Honestly, no. Changing human behavior takes time. But these quality checks will stick around, offering lasting benefits and efficiencies. Now, let's dig into how we pulled it off. Ready for the dog journey?

How did we do it?

As mentioned before, the success in launching new processes shouldn’t be taken for granted. We considered the following steps for launch.

Finding the right coaches

The need to form a robust team in alignment with the program's strategy cannot be understated. We firmly believe that, for the success of this initiative, it's crucial to involve the expertise of product management, product design, and engineering – what we playfully refer to as 'the three amigos.' In this instance, product operations assumed a central role in driving the initiative forward.

Here’s your checklist of teammates:

  • A product operations individual to facilitate the conversation
  • A product management representative to bring the product experience and expertise
  • A product design representative to cater for how visual feedback should be handled
  • An engineering partner to secure the right integration with existing development processes

Agreeing on what food to eat

In simpler terms, it's about pinpointing the appropriate scope. We need to strike a balance—not too expansive, as that might dilute the feedback, and not too confined, as that might limit exploration. Hence, for our program, we made a clear decision: we'd exclusively test the products related to our Marketplace platforms, namely, and the mobile applications available on both iOS and Android.

Setting up the room conditions

Alright, we've got the metaphorical food, identified by our coaches. Now, let's focus on setting the stage for the program's development. How do we create optimal conditions, and what's the game plan for dealing with the outcomes? In essence, the key lies in establishing a robust process capable of effectively managing all the feedback we collect – undoubtedly the program's linchpin.

There are some important aspects for you to consider:

  • The bowls are clean: Before diving into prioritizing dogfooding feedback, it's crucial to establish a baseline of the bugs we've historically reported. This step is instrumental in minimizing unnecessary noise and discerning what issues originated specifically from this channel.
  • You know how to clean up the mess: which is caused by the “food-eating” process. In our case, we have created a process that integrates 3 tools:
    • Slack: a public channel that uses Workflows to collect feedback from
    • Google Sheets: a place to aggregate and where product managers (PM) will analyze all submissions
    • JIRA: the tool we use to integrate all the relevant dogfooding tickets with engineering teams after PM screening
  • Dogfood the dog food: Just like a cook wouldn't serve a dish without tasting it first, we follow a similar principle in our program. Seize the chance to test the process within the team before the official launch, conducting an initial round of feedback.

Letting the dogs out

Who, who, who, who! 

Crafting a launch and communication plan is a necessity. In our strategy, we opted to kick off the process with our product team initially, with plans to expand to additional teams in subsequent cycles.

As such, we planned 3 steps:

  • Documenting the whole process on Confluence for the sake of knowledge transfer
  • Setting up a cool filled-with-dog-puns deck to be presented at an All Hands session
  • Inviting everyone to the playground (aka the Slack channel) and unleashing the power of dogfooding

Here’s a nugget of wisdom: never underestimate the influence of your product individuals. Ensuring their enthusiastic involvement is the key to ensuring the success of your program.

What we've learned

Seven months in, 100+ submissions deep, and we've got some insights to share. Let's skip the product specifics and dive into the program's wisdom.

  1. Don’t underestimate the power of leadership alignment: Seriously, don't skip this step. Leaders are your program's sponsors and champions of product quality. Realign as needed because a top-notch experience is the name of the game.
  2. Understand the context and adapt the gameplay: For any dogfooding process to thrive, the entry barrier must be exceptionally low. To achieve this, delve into the intricacies of your team's current operations, such as sprint cycles, tools, and workload, and tailor your process accordingly. Embrace pragmatism and realism, recognizing that each team is unique.
  3. Game rules might need adaptation along the way: To be fair, it's entirely acceptable! Life evolves, business priorities shift, and team focus may waver. Adaptations to the program, like adjusting the scope or refining tools, might be necessary. Nevertheless, the core focus should persist: dogfooding is a powerful way to identify issues early in the process. After all, practice makes paw-fection!
  4. Product individuals may not be comfortable with raising feedback on their own work: As it may sound self-sabotaging, team members might not feel comfortable snooping around the product. It is important to create an open and honest culture where constructive criticism is encouraged. The team sets the pace and the success of this program.
  5. Think about measuring success: Connect the dots for leadership; they'll want numbers. And you’ll want now and then to check the vital signs of your program. Ensure success metrics are there from scratch; examples can be the number of feedback given, the number of fixes performed, how many users have provided feedback each month, etc. 
  6. The play will stop being fun at some point: New things lose their sparkle eventually. Plan a set of challenges/scavenger hunts across the product journeys to be shared with your team or bug-bashing events, and you can choose when to activate them and gamify the program a bit!
  7. Community makes perfection: Dogfooding shouldn't just be a habit; it should be the cornerstone of a product community's ethos. Make it a focal point, and your program will naturally sustain itself. Fall short of achieving this, and you'll find yourself perpetually trying to herd the team to keep everyone on the same page.


In wrapping up our dogfooding expedition, we recognize that not every organization or team may fetch the same benefits from this approach. Understanding the nuances of your reality and the need for adaptation and flexibility based on the nature of your product and team is crucial.

If you choose to embark on the dogfooding journey, it's more than just a taste test—it's a chance to truly empathize with customers. So, whether you're a seasoned dog in the industry or a new 'paw-spective' player, remember, the insights gained from dogfooding could be the treat that keeps your customer experience tail-waggingly delightful.