OnStartups

The Growth Hacker's Dilemma: Process vs. Tactics

Posted by Dharmesh Shah on August 27, 2013 in guest marketing 23 Comments

After conducting nearly 100 interviews with some of the world’s best growth hackers on Growth Hacker TV, I have become keenly aware of a certain tension that is in the growth hacking ecosystem. Some growth hackers choose to emphasize the process of growth hacking while others choose to see growth hacking as a set of tactics that can be applied to various scenarios.


First, let me define the growth hacker’s process. There is no one single agreed upon order of operations, but a growth hacker’s process is based loosely on the scientific method. If you can remember high school, the scientific method is basically the following:


  1. Question - Why do visitors leave our registration flow after the first page?

  2. Hypothesis - They might be leaving because page two has too many form fields present and this scares them away.

  3. Prediction - If we have more registration pages, but less form fields on each page, then our completed registrations will increase in statistically significant ways that could not be the product of chance.

  4. Testing - For the first 2 weeks of September we will run an A/B test, showing 50% of new visitors our current registration flow, and showing the other 50% our new registration flow which increases the number pages but decreases the fields per page.

  5. Analysis - The results show that our new registration flow had 27% more completions than our current registration flow, and this is statistically significant enough to conclude that we should implement our new registration flow.


Here is where things get interesting. Some startups will actually use this scientific method (or something similar) as a means of gaining insights about their product, thereby enabling them to make progress. Others, however, will not have a rigorous process like that listed above, but they will instead use the results of other people’s experiments. Put another way, some startups have a process, other startups just implement the tactics (best practices) that are the results of someone else’s process. If someone read about the above experiment on Quora then they might adapt their registration flow without a scientific process in place to support such a move.scale tradeoffs


The question is, which kind of startup should be applauded and which should be reprimanded? It might seem obvious to celebrate the rigors of the scientific method and side with any startup that uses such a process. However, I think there is a case to be made for both kinds of companies. Obviously, if someone doesn’t run the experiments then we will never arrive at the tactics in the first place. The tactics are the byproduct of someone’s hard work and that should be appreciated, but think about how the scientific community actually operates. The scientific method is a tool that serves the entire scientific community, and the results of that tool are often fair game for the community. Scientists don’t expect each other to run every relevant experiment for their personal endeavors. Newton said, “If I have seen further it is by standing on the shoulders of giants.” Why can’t a startup simply use the results, as discovered by their fellow entrepreneurs in lab coats, as a benefit of the community?


The truth is, there are pros to both ways of thinking, which I’ll list below, but I don’t want us to view growth hacking as only a process or only a set of tactics and simultaneously miss the community aspect of our enterprise. Here is how I see things:


  • Pros of Process Oriented Startups

    • Without process oriented startups we would have no tactics.

    • They are able to find new growth hacks when old ones cease to work.

    • By understanding the process, their implementation of any given tactic will be more nuanced and effective.

    • They are more self sustaining, able to use the community, but not be entirely dependent on it.

  • Pros of Tactic Oriented Thinking

    • Allows smaller startups, with less funding, to implement tactics very cheaply.

    • It is an entry point into growth hacking which is more accessible than experimentation, even though it might lead to experimentation later on.

    • Not every experiment needs to be ran by everyone. Some best practices are near universal and can just be applied. We all do this to some degree whether we realize it or not.


So, what is the answer to the dilemma? Is growth hacking a process or a set of tactics? Well, both, and here is what that means practically. If you are in an organization that has a growth hacking process in place then see yourself as a part of a larger community. We are grateful for your work, but you don’t need to be pompous about your place in the universe. Share what you find, grow our collective knowledge base, and understand that not every company will imitate you, and that’s ok. If you are in a non-process oriented startup that is still trying to use growth hacking principles then be extremely appreciative of the companies that are supplying these best practices, and consider creating your own process so that you can give back to the community as much as you take from it.


Startups aren’t going anywhere, and growth hacking is here to stay as a robust methodology for growing them. Whether you are in the lab, or reading the research paper that was spawned from someone else’s lab, understand that this is a community, not a zero sum game.


This article was a guest post by Bronson Taylor who is the host and co-founder of Growth Hacker TV, where the experts on startup growth reveal their secrets.