How to Design a Growth Process that Scales

I always like to say that Product, Team and Process are the 3 pillars of Growth.

It’s rare that a company manages to ‘copy/paste’ another company’s Growth Hack with similar success. Not only is every market and industry different, but every company is different too; to such an extent that even direct competitors would often struggle to successfully copy each others’ ideas.

True Growth comes from having an amazing Product, a multi-disciplinary Team and a Process that allows this Team to test, execute and scale ideas and strategies for Growth.

In this article I’d like to focus on the Process, as it is something that many companies struggle to implement successfully.

What is a Growth Process?

If you either work or are interested in Growth, chances are you are familiar with the concept of a ‘Growth Process’. Over the past few years, thought leaders have developed a number of different ‘versions’ to this kind of process.

At its core, the concept of a ‘Growth Process’ usually refers to a process for Experimentation. Born out of the Lean movement, the validation of assumptions through effective experimentation is a core unique feature of Growth Hacking. One of the most known versions of such a process is the one designed by (what a surprise) Sean Ellis, the ‘founding father’ of Growth Hacking and CEO of

Sean Ellis’ Process:

Sean Ellis’ Growth Process

As is the case with much of Sean Ellis’ work, his process has formed the basis of most Growth Processes out there. You begin with an ‘Objective’ (which some might call the ‘North Star Metric’ or the ‘One Metric that Matters’) and then you come up with ‘Ideas’ to reach it. Next, you ‘Prioritize’ these ideas, ‘Test’ them and finally ‘Analyze’ the results.

Many others have gone on to create their own Growth Processes, inspired by that of Sean Ellis. When I was working as a Lead Speaker and Growth Hacker at Growth Tribe, we taught our own version of this Process, developed by David Arnoux.

David Arnoux’s Process:

Growth Tribe’s ‘GROWS Process’

You can see the similarities to Sean Ellis’ process (Ideation, Prioritization, Testing, Analysis), but David’s process underlines the importance of designing your experiments before executing them (‘Outline Experiments’). When you’re testing things, it’s very important to agree how you’re going to test something before you go and do it.

Ryan Gum, who now does Growth at HubSpot, also shared his Growth Process with ConversionXL.

Ryan Gum’s Process:

Again, you can see the similarities, especially in the Process’ circular and iterative design, but there are some differences. Sean Ellis has a central ‘Objective’ in his Process. Growth Tribe use the ‘One Metric that Matters’. Ryan Gum, however, does not focus on a single goal, but rather refers to ‘goals’ in plural. I’ll explain why this is important in a bit.

There is also one interesting addition to this process, which is the ‘Optimize’ step. Optimization is vital when it comes to scaling ideas, something that most processes tend to overlook…

The Problem with (Experimental) Growth Processes

The processes above offer a solid basis for experimentation, which is why I based our own experimental process on many of the principles of those above.

Recruitee’s Growth Process:

Recruitee’s Experimental Process

However, Growth is about more than just experimenting. I’ve mentioned before how I’ve seen too many teams experiment for the sake of experimentation, always testing new things at the expense of scaling what actually works.

Rapid Experimentation may be the key to Growth, but scaling successes is vital.

There’s a difference between saying that failure is ok in order to encourage creativity and actually trying to come up with crazy ideas that are likely to fail. The whole point of ‘Prioritization’ is to avoid failure and hopefully succeed with the least amount of tries.

As a result, I’ve chosen to drop the circular design. Yes, a Growth Team should test new things and should not be stopped by a fear of failure. But they should also always use their best judgement and only iterate on experiments if their previous tests failed.

In our Experimental Process, we start by setting a goal. Brian Balfour recently wrote about how the North Star Metric and One Metric That Matters can be misleading. I tend to agree. Anyone working at a startup or scale-up will tell you that it’s hard, or even impossible, to imagine setting one metric or goal that everyone will exclusively focus on (unless it is an overarching long-term goal tied to your vision). That’s not how startups work. We’re always in the trenches, we’re often putting out fires, we’re all doing a bit of everything and we’re all fighting for every inch of ground on the competitive battlefield. Instead, we choose to set more general goals, for which there can be a number of input and output metrics and around which we come up with ideas to experiment on.

The rest of the process is pretty standard, except that I’ve put special emphasis on scaling the successes. If your experiment succeeds, you need to scale it and then constantly optimize to get the most out of it.

But what happens once an experiment succeeds? What about ideas that cannot be tested in a short period of time?

A Process for Scalability

What is a Growth Experiment? It’s a pretty broad concept. On the one hand, it can be as clear as saying “I will test LinkedIn Text Ads”. On the other, it could be more along the lines of “I will test the assumption that pain point X has the biggest impact on users’ decision to purchase so that I can incorporate that into my next influencer campaign”.

The former experiment is of course the fastest and easiest to manage. It is often also the most satisfying because you get to see real Growth-inducing results in a short amount of time. However, as your business scales and you begin to figure out what does and doesn’t work, you will find your experiments drifting more and more towards the latter; after all, even with more channels popping up every day, there is only a finite number of viable channels to test. Whilst that might be less tangible and satisfying, it is all the more important.

Every successful experiment should be scaled and every long term strategy should rest upon assumptions that have been validated through experimentation.

For example, say you want to test a brand new feature in your product. Chances are, this won’t be something you can build and test in just a couple of weeks as it will require multiple people, significant resources and a good deal of opportunity cost. Still, pretty much every long term idea or strategy rests upon a number of key assumptions, as mentioned above. The real object of Experimentation will most likely be those core assumptions that constitute your bigger ideas, which you will be able to test in a number of ways and assess with both hard and soft data as well as early indicators.

So what about a process for the big ideas and strategies? As your company grows, you will find yourself increasingly working on bigger ideas and more long term strategies. This is completely normal. In these instances, the Experimental Process is just one step within the wider Process of a bigger Project.

This is something I’ve found to be missing in most companies. If you have a process for experimentation, then surely you should have a Process for scaling experiments, i.e. what we call ‘Projects’ here at Recruitee.

Recruitee’s Project Process:

This is a process we developed to execute our long term ideas as efficiently as possible. Sometimes this process directly follows a successful experiment (the ‘scaling’ part of our Experimental process above) and sometimes the strategy idea comes first and then we experiment on the assumptions.

First, we Outline the Project, breaking it down into individual tasks, which we assign to members of the Team with clear and realistic deadlines for each task. Every task is ‘owned’ by someone in the Team, meaning that they don’t necessarily have to execute on it, but they need ‘own’ it and make sure it gets done on time.

Next, all Team members execute or coordinate the execution of their tasks. Throughout and beyond this process, everything is tested and verified to ensure everything is working and all assumptions are validated.

Once the project is fully executed, we analyze the results and continuously iterate and optimize to get the most out of it.

Some Final Thoughts on Process

Of course, most Processes that are beautiful on paper do not look as clean cut in practice. This is fine. The whole point of being a lean company is to be able to adapt and mutate to be as effective as possible.

In my experience, companies with little structure should only be strict about their Process in the early days. Once the Team matures and everyone owns the Process, it is fine to deviate, skip steps, improvise and adapt according to the situation.

With that in mind, all of the Processes mentioned in this article are valuable for any Team looking to scale a business. The biggest issues I’ve seen companies have do not come from limitations within their ‘on paper process’ but rather in the adaptation and execution of the Process.

This is the first of 3 articles on the ‘3 pillars of Growth’. Check out the second article on ‘Team’ and stay tuned for the last article on ‘Product’.

Leave a Reply

Your email address will not be published.