A guide to dogfooding

How do you create outstanding products?

If you ask around, many people will tell you that you need to understand your customers and their situation and empathize with them. Once you’ve done that, you can create products based on your unique understanding of your users. While I don’t disagree entirely, there are better ways to approach this.

For instance, what if you become your own customer? Yes, it’s important to understand your customers, but it’s also important for you to understand the product that you create so that you see the value it provides. In the PM world this is called dogfooding.

From this article, you’ll learn what dogfooding is, how to apply it, and how to understand the results of your sessions so that you can make more informed product decisions.

What is dogfooding?

Dogfooding refers to a situation where a product team uses their own product or service before releasing it to the market as a test case.

Let me share a story to illustrate why dogfooding is highly important.

My favorite restaurant is in a small town in Brazil called “Tio Giuseppe.” What amazes me about this restaurant is the quality and simplicity of its food. When I lived in Brazil, I visited this restaurant whenever I could, and once I talked to the chef, it amazed me what he shared.

For years, the restaurant struggled to thrive. The chef didn’t understand why as they offered many beloved dishes. And yet, customers would rarely return. To get to the root of this, the chef decided to have dinner there and go through the menu, and the result was anything but good.

He asked the other cooks to do the same and the reactions revealed their real problem — the food wasn’t as good as they imagined. The chef closed the restaurant for two weeks and told the other cooks, “We will create a menu that we love. First, we must love our food, and then customers will be pleased.”

Over 200k developers and product managers use LogRocket to create better digital experiences

Learn more →

They ran many experiments, removed the menu, and started serving only three dishes: lasagna, spaghetti with organic meatballs, and fish of the day. After that, customers started recommending the restaurant. They loved the new concept and the quality of the food.

How can you apply dogfooding in tech?

Dogfooding isn’t about faking customer behavior to prove yourself right. It’s about becoming your customer and understanding how your product improves your life.

Let’s look at an example that Itamar Gilad, former Google Product Manager, shares in his book “Evidence Guided:” it’s a story about Gmail. Nowadays, it’s natural for you to receive your emails automatically categorized and take it for granted.

The following image reflects my Gmail account:

One unread email, a few hundred social emails, and another few thousand promotions, which I ignore. I love this functionality because it helps me avoid distractions and focus on the emails I care about. Yet, how that came to exist is quite interesting.

Itamar was one of the Gmail PMs; he shared that back then, emails were crowded, and he noticed people spending considerable time sorting them out. Then, he imagined, “What if we could do it automatically?” That question triggered a series of experiments.

To automatically sort the mailbox, Itamar aligned with the team and agreed to use the new version so they could experience how that would play out. They all used Gmail and it was natural for them to evaluate the new feature. In the beginning, it was helpful but not as much as they had imagined. So they kept improving it.

Once the functionality reached A-level, the team benefited from it every day, and they decided to get customers to use it. At that stage, the functionality was already mature and it was easier to deliver value to customers.

The role of dogfooding in product testing and development

One of the major benefits of dogfooding is the ability to step back. When teams don’t practice dogfooding, they risk incorrectly assuming users are wrong about something because they don’t understand how a feature actually functions. This is detrimental for both sides.

Dogfooding helps you understand customers better because you start wearing their shoes, accelerating value creation and reducing friction. Yet, it doesn’t mean you can stop talking to customers altogether. Dogfooding isn’t exclusive — it’s complementary. The closer you are to your customers, the better.

Steps for running dogfooding sessions

How do you run dogfooding sessions? It differs from situation, but let me give you an overview:

• Define — Make it clear what you’re trying out and which job it solves for. Remember, it’s about delivering value, not testing the feature
• Testers — Agree on who will use the product. Somewhere between three to ten employees would be a good starting point
• Time — Establish a window you want to test. This can vary from a single session to days or weeks. I recommend keeping it short to collect insights faster
• Resonance — Let the testers share how the created feature helped them achieve the desired job in real life
• Pattern — Strive to map the identified patterns during the session
• Rinse and repeat — Improve the product based on your learning. You may opt to stop dogfooding if you realize the testers achieved the goal in a satisfactory way. Otherwise, you improve until you’re ready to involve customers

Understanding dogfooding in the context of testing

Dogfooding is a qualitative test. It’ll help you know how useful the product or service is. Although it may help you identify bugs and inconsistencies, it’s not a scalable testing method to keep your quality standards high.

Dogfooding helps you identify what drives value and iterates until it’s intuitive and easy to use. Before releasing features to your audience, you’ll benefit from adding test methods like smoke testing, regression testing, and sanity testing.

Key takeaways

If you strive to find ways to determine how your customer interacts with your product, you’ll deliver highly valuable products sooner. One of the best ways to do this is to become your customer. When you do this, you may learn why customers struggle with, or even decide to drop your product in favor of something else.

Although the benefits of dogfooding are great, make sure you implement it alongside other testing practices. Smoke, regression, and sanity tests are still relevant and provide complementary data to give you a more robust sense of the health of your product. It’s never a good idea to rely on a single data point.

Featured image source: IconScout

LogRocket generates product insights that lead to meaningful action

LogRocket identifies friction points in the user experience so you can make informed decisions about product and design changes that must happen to hit your goals.

With LogRocket, you can understand the scope of the issues affecting your product and prioritize the changes that need to be made. LogRocket simplifies workflows by allowing Engineering, Product, UX, and Design teams to work from the same data as you, eliminating any confusion about what needs to be done.

Get your teams on the same page — try LogRocket today.