How to Be Systematic When Testing Marketing Hypothesesby@refocus
1,498 reads
1,498 reads

How to Be Systematic When Testing Marketing Hypotheses

by Roman Kumar VyasSeptember 9th, 2022
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

Given an integer `n` is a power of four, if there exists an integer such that `x` such that`n == 4x` The easiest way how to solve this problem is to divide N on 4 and that is it. We will iterate again and again till our number is bigger than 1. Author: Sergei Golitsynlink: < an integer `n` return `true` if it is a power of four. Otherwise, return`false`

People Mentioned

Mention Thumbnail

Companies Mentioned

Mention Thumbnail
Mention Thumbnail
featured image - How to Be Systematic When Testing Marketing Hypotheses
Roman Kumar Vyas HackerNoon profile picture

Since the company I founded, Refocus entered the Philippines market in January, a lot has happened for me. We conducted many types of research and CustDevs, tested tons of ad creatives —  and even changed the product completely. As a result, our revenue for 6 months has reached 1.5$ million. It became possible for a whole set of reasons: developed product, constant qualitative and quantitative research, and of course, marketing strategy.

Here is the system we created the marketing to make it possible in 4 steps.

I. Analysis of the demand

Southeast Asia market experiences a shortage of IT specialists: currently, the companies there are estimated to lack between 600,000 and 1.2 million professionals and this number is predicted to reach 4.3 million by 2030. And that's the problem we're determined to solve through our course in Digital marketing that showed excellent metrics in other countries.

When we just started creating the program, we thought that it will be targeted only at people with zero knowledge. However, market research showed that our audience actually consisted of 2 segments:

  • Reskillers —  people who want to change jobs and learn digital professions from scratch
  • Up-skillers —  those who want to improve their skills.

So, we started testing 3-5 pitches for each segment but none of them were appealing and the conversion rate was pretty low. 

It was a good lesson that the product which was in demand in other countries can be not that wanted in the new market. And that’s ok! You just have to dive deeper into the research and find out what people needed — and of course, that’s what we did:

  1. We generated hypotheses about which professions were likely to attract attention. As the philippine Ed-Tech market is not well-developed yet, we built our hypotheses on our own experience in other countries, our competitors in other countries, and, of course, on the research dedicated to the lack of digital professionals. Among the list we created were: Web Developer, Business Manager, Data Analyst, Project Manager, and so on.
  2. Running test ad campaigns to see if people are actually interested in studying those professions. We created a couple of ad creatives for each possible product and checked CPL and CR.
  3. Then we calculated the results and repeated the testing to reassure that everything was correct.

And —  data analytics appeared to be both the most in-demand and desirable. Indeed, according to, companies experience the highest shortage in Data Analytics than any other IT specialists, with the demand for it growing by 131% during the last 6 years.

So, we started working on another course.

II. Building the funnel

Experience in other countries has shown that the best-working lead-attraction methodology in EdTech consists of two main ways of their acquisition. Combined together typically bring us 90% of the leads; they are:

  • Direct funnel. Here we include targeting in Facebook and Google - each forms its’ 30% of traffic
  • Webinar funnel. We conduct free webinars where we share some basics and then suggest the best offer for the course. From it, we get another 30%.

Of course, one shouldn’t forget about SMM and other branding tools. But the two funnels mentioned above indeed can generate 90% of the leads if they are built correctly.

1. Direct funnel

Similarly with the Digital Marketer course, we started to create several pitches for each segment of the audience. The only thing that changed was that we already knew that the course was attractive to the potential students: we didn’t know which triggers would be the most appealing.

From our Customer Development we knew that a person usually joins a course to find the job that will help them improve their lifestyle: work remotely, travel, spend more time with their family. So, that was what we started testing. Based on the conversion rate of the ad creatives, we determined 3 best-working pitches.

  • promise to learn the material “from scratch to occupation” 
  • a competitive portfolio during the course
  • the fact that the skills one acquires are highly marketable. 

In contrast, the “earn to travel” trigger attracted many leads but showed low CR, while naming the price directly just seemed to be unappealing.

2. Webinars

Pretty the same here: we needed to determine what were the most interesting topics that brought the most leads and showed the most conversion into the sale afterward. 

It turned out that 2 factors were determining the success of the webinar: the choice of the speaker and the topic. For example, the same topic could be completely ineffective with one speaker but give excellent results with another. Or similarly, the same speaker talking about different themes could attract many sales afterward or literally none.

Now we finally were ready to take the next step!

III. Systematization Of The Results 

As soon as the product was determined correctly, revenue grew and so did the marketing budget. Now instead of around 12 advertisement campaigns per month that we had for the Marketer course, we were able to have 50! This shift is vitally important: now we can constantly improve our creative materials without harming the unit economy and thus attract more clients cheaply and more effectively.

To keep track of it we’re using the tool named communicative map which we informally call “a creative machine” 

Basically, it’s the table that includes just everything the team should know about each creative tested and helps to organize the work together with copywriters and designers: 

Then we include some technical detail: who is responsible for the certain design, how many of the creatives is needed, the link to them, and so on. You can choose for yourself!

And the last part of the table is dedicated to key metrics.

So, we always see that one or another creative is having good CR and CTL - and can improve it even more!

IV. Keeping testing.

As soon as all the intermediate results are put into the table and analyzed, one can take the best practices — and proceed with testing. In this situation, we were simultaneously working on improving both ways of lead acquisition.

1. Direct funnel. Here were focusing on the creatives and changing:

  • The form of the material —  text or visual. If your business is not related to the beauty industry or a store of goods where visuals are very important, you are better off testing the messaging. You can do this even on a simple neutral background. Practice shows that in the service industry, the text works better than the image.
  • Format of the creatives. We tried to activate the various popular formats and see the response. We checked that in the education industry, for example, such formats work well: a pitch from the founder, student feedback, avoiding mistakes, solving the audience's pain, the expertise of the course instructor, etc.).
  • The amount of information on the creative. For example, during the test, we found that ads with the date of the event show worse results, and it's better not to write them. Actually, it's better not to overload the creatives with information.

2. Webinar funnel. In this case, we were focusing not only on the advertisement but also on the content of the webinars:

  • The balance between sales and knowledge, given during the webinar. Logically, people value when indeed unique information is being shared. If the part with the sales is too long, we got a lot of messages that the whole webinar was a waste of time. 
  • The amount of personal story. It turned out that listeners value such stories a lot, especially, when a speaker is a Filipino person with a similar background who managed to reach what they’re dreaming about. So, the webinars where the speaker was describing their experiences in detail, showed the best conversion rate.
  • Time of the webinar. We tried to change the time and days to see when we can get the best attendance. It turned out that the evenings of the weekdays were much more often visited than the weekends when people prefer not to think about education.
  • The number of messages sent to the registered client. In this case, we were seeking balance: not overloading the clients with spamming messages but not letting them forget. Typically, we send one upon registration, another a day before the webinar, and one more 15 minutes before it.

IV. Further work together with other departments

No matter how hard the marketing team works, you can't get the proper results without interconnection between different departments. For example, one of our best decisions was to deepen communication between marketing and sales. The latter communicates with the leads directly, so know their pains like nobody else. For example, they said to us that clients really believe in FOMO triggers and value the knowledge they could get for free.

That’s why this creative worked just perfectly: we say that the webinar is free only for some time and that they will not only get knowledge but also receive a certificate for free.

To Summarize and Motivate

The system is everything! Don't think the good metrics are luck - in fact, they are determined by pretty simple and logical things:

  •  proper market research that will help to create a valuable product
  •  testing best practices in creatives and most importantly analyzing the result
  •  applying best practices after the tests and constantly improving them
  •  learning insights from other departments

And of course, it's impossible without patience. Just keep going and your startup will stand out one day!