Activation Hypothesis to influence number of orders

------First, I must briefly introduce the product. I haven't developed a million products YET, but there are a few of them. One of these is an e-commerce platform created inside the biggest marketplace from Moldova. Sounds complicated. Well, it is complicated indeed.

------Everyone tells me not to be afraid of mistakes, to learn from them, and to move forward. So I've decided that my favorite column will be: Ina's f*ckups, or how to fail a hypothesis from start to finish, or the importance of Analytics, or call it however you want. Let's go!

Come on guys, we can't continue to receive 3 orders per day!

Step 1: The problem and hypothesis preparation

------We logged into Google Analytics and took one more look at the beautiful dashboards. Millions of visitors, great, let's go home! But wait, there is a tiny problem. Even though people are visiting our website, we have a very low conversion in successfully delivered orders.
------Let's get to work. I figured, to increase the number of orders, we need to increase the number of product views. Great, at least we know what metrics we want to influence. Now we have to figure out how to do that? I guess I have worked just with the first idea that popped into my head.
------The hypothesis: If we add a button in the navigation bar on the mobile version of the website that will filter items with delivery in subcategory X, then we will increase the number of orders by 6%. Nothing foretold troubles.

Lesson Learned: Don't choose the main metric from another step of the pirate funnel

We knew we didn't have high conversion in orders, but we didn't analyze where users were getting lost. Maybe they view the product, but the order form is too complex, or maybe the delivery is too expensive.

No help from Development Team? No problem, Optimize's got your back.
Step 2: Running the hypothesis
SS
------The hypothesis is formulated, now we need to test it as quickly as possible and validate the idea. I started with prototyping the experiment, creating 3 versions of the design for that button. One of which was in the shape of a switch. The designer confirmed the last alternative with the switch and all I have left to do is to "code" in Optimize.
------It took me a couple of hours to execute. Thanks Google for making such a tool! I've created a personalization test in a subcategory on mobile version of the website, in Romanian and Russian language. I've launched the experiment and impatiently waited for the results.

Exciting results. Let's do the same in all subcategories

Step 3: Results analysis and hypothesis scaling

------After 2 weeks of running the experiment, it was time to analyze the results. I was looking at the number of orders placed before the experiment, and compared it to the number of orders during the experiment. The growth is noticeable, about 40-50% on average. But the main question remains: did my experiment influence this increase or did other factors? Another f*ck up.
------I decided to scale the experiment on all existing subcategories in the product. I reconfirmed the final design and discussed this task with developers. But integrating such a change implies modifying the whole system, including desktop version. To avoid making the task more complicated and changing the whole filtering system, we made this change only in the mobile version for the products list page and for shop's profile page.

Analyze the entire system and let analytics become your best friend

In the end an experiment that was supposed to increase the number of orders, has actually complicated the Customer Journey Map. The user experience suffered, and I' m not even sure that this experiment influenced the core metrics. This hypothesis taught me to radically change the process of Growth Hacking, analytics and teamwork.

------First of all, learning from mistakes. Secondly, I will describe these lessons in the following articles. I invite you to follow my progress in creating new hypotheses, and maybe subscribe, and maybe buy me a house, or maybe a yacht. Keep in touch.