The term 'dogfooding' has been around for years but realistically how do you do it and why is it desirable?
Eating your dog food
To be able to practice the art of ‘eating your own dog food’ requires you to have a product that you can use yourself, and by doing so, test and improve upon it.
In the years I – James Kelway – spent on the agency side of digital product design, most clients we had either did not have the environment of learning to enable this to occur or the collaborative culture required to deliver products quickly enough.
Thats not to say companies cannot do it, but there are some important aspects that should to be considered if you are to employ this highly valuable form of customer feedback for the development of your product.
Ensure you own the customer ecosystem
This means you need to have both the delivery mechanism to supply the product and the quality measures and benchmarks in place to guarantee it is being delivered.
In our case we built a grant management system on the back of 10 years of experience in working with grant seekers and also having a product that handles 40,000 grant seekers online, in peak periods, every month.
Having the data from a live product, means we can design a system for both applicants and grant managers. The platform gives us rich insights that avoids us being feature-led.
We believe a generic, best practice way of handling applicants is what time-poor smaller foundations need. They need accessible, affordable and professional tools that enables application handling and resolution in an entirely digital, compliant and secure environment.
“Being aware of your limitations is a strength, and finding a partner who knows more than the product team about certain critical subject matters is essential.”
Only make what is valuable
If you are using the product you make, then you need to become the customer. In our case we created a grant for students and launched it for a short time span of one week through our Legatbogen online database of grants.
Legatbogen is a free service with over 10,000 foundations indexed that provides the largest source of grants for applicants in Denmark. We created our foundation and grant through our new product and within 48 hours we had over 150 applicants who had come via the online grant database.
Though all parts of the product had been through numerous validations and user tests, this was the best signal that the product was fulfilling its aim. With real applicants, and with only a few small issues being reported, we could justify proceeding with the release.
Prioritise every day
There is always a huge backlog of features and ideas as you progress but try and ensure a core of a product exists that fulfils the job to be done.
Every actor within a system needs to get something accomplished and each actor’s needs will create their own set of functionalities that couldbe designed and built. Letting this list cloud the vision you may want for your product is the constant balancing act that the team (and company) needs to pay attention to.
We hold weekly meetings to discuss the impacts of decisions and ensure we have the forum to air concerns if we think it will lead us in a direction we do not want to go in. Priorities have consequences and that inevitably leads to compromises for somebody, somewhere.
Being alert to the impact of decisions early on creates a clear product roadmap and the right strategy to follow. Also saying no to features gives clarity and inevitably a closer product-market fit to the customers who we want to serve.
Find a true development partner
Being aware of your limitations is a strength, and finding a partner who knows more than the product team about certain critical subject matters is essential.
Make your partner a core part of the creation of the product. That way you embed their perspectives as part of the solution. This will distinguish the product by having their expertise baked in, and when it’s good enough they will also use the tools alongside you.
Dogfooding isn’t easy to do well, however it is easy to do it badly
It may be impossible for some situations and many companies do it without having the key ingredients in place. We can’t all be Google or Microsoft after all, and many have criticized it for being applied in the wrong contexts.
Launching and using a living product yourself results in feedback that is immediate. It requires action and sometimes the need to do a fix is urgent. It forces many more releases and an agility of a team to respond to live events, so the turn-around needs to be hours or days, not weeks or months. Some companies will find it very difficult to respond in time and that is why it is difficult for companies to act in this way.
But above all, there must be a specific culture for this type of product development to take place in. It needs to be successful but also enjoyable too. It needs continuous releases, not just phases of development work. Even sprints can be difficult to plan as priorities change within that time.
To our customers, the product should continually improve in small, complete increments. Some things take minutes and some take days, but for a customer the effort is indistinguishable.
Keeping what is valuable to the customer at the product’s core and shipping as often as possible, will ensure that the product’s value is realised early. That is why if you feel that ‘dogfooding’ can be done, and have the ingredients in the recipe above, then you should give it a try.