3. Choosing Channel Mix and Finding Product-Channel Fit
- Arindam Paul Founding Member and CBO, Atomberg
In this snippet, Arindam introduces a 2X2 framework to think of acquisition channels and consider their tradeoffs while building your channel mix.
If you examine all acquisition channels, similar to how we used to organize everything in a two-by-two matrix, there are some channels that provide immediate results, while others take time to yield results. Some channels are scalable, while others are not.
If you plot on a two by two matrix, I think the non scalable ones have a very small use case in the journey of an organization. But among the scalable channels, there are two types of channels — channels that are scalable and have quick results, and channels that are scalable and have slow results.
Scalable and Quick channels often have a diminishing returns curve. You scale linearly up to a point, and after that, it starts to flatten out. It becomes increasingly expensive to acquire incremental customers. This is why you can't scale these channels beyond a point.
For example, someone might suggest, “Keep acquiring customers faster and maintain the same budget.” However, it’s not possible due to this phenomenon.
For instance, even in ATL (Above the Line) campaigns, if I want to reach 55% or 60% of consumers, it’s more efficient. But if I aim for 90% of the audience, diminishing returns apply.
Then there's another type of returns curve known as the "S curve." This curve starts slowly but yields results over time. These are channels that are scalable and slow. Consider things like content or SEO. When you invest in them for the first two or three years, you may not see significant results. However, if you keep investing, at some point, you'll experience a significant and rapid spike in growth.
That’s why most people give up on this channel, saying it doesn’t work for them.
The same applies to ATL (Above The Line) advertising. Many people believe that if they run a TV ad today, their brand awareness and other brand metrics will immediately start rising, but it doesn’t work that way.
It takes a lot of time, spending, and effort. It’s crucial not to stop before you reach that scale. It’s essential to understand that once you hit a certain scale, you need to activate other channels; otherwise, your Customer Acquisition Cost (CAC) will become ineffective.
That is why many brands find it very difficult to grow from hundred crore to a thousand crore.
You see many brands reaching the 100 crore mark because until 100 crore, acquisition works linearly. The more you spend, the more you scale. However, beyond a point, diminishing returns set in and growth becomes more challenging.
For example: In the case of content, having a substantial following, say, reaching 1 billion followers, makes it much easier to grow by another 100,000 than growing from 0 to 100.
However, in the context of performance marketing, the dynamics are quite the opposite. Once you hit a certain scale, and you want to grow from there, you have to spend significantly more than you’ve spent before.
Below, Arindam explains how channel mix evolves as the company grows, focusing on unscalable channels in the early stages and scalable channels at later stages.
You need to start your acquisition journey with unscalable channels. These are organic channels that already exist and are not paid or owned. It could be some WhatsApp group, a Telegram channel, a Reddit thread, some niche blogs, or some niche websites where you piggyback on top of that audience and get an initial, early scale of consumers.
The whole objective of using unscalable channels is to test product-market fit. Once you do that, the next stage is to start testing the scalable channels.
Testing scalable channels is very important because acquisition channels also have to fit with the product. You need to figure out which channel is the most effective for your specific product. For example, medical devices might not work on Facebook due to regulations. Your goal here is to identify the most suitable channel. As you do this, you can gradually phase out the unscalable channels.
In our case, we chose Facebook, which might seem counterintuitive given that we’re selling fans, which aren’t typically an impulse purchase. However, from 2017 – 2019, our lookalike audience campaigns on Facebook started doing well. We captured data from around 5,000 to 10,000 consumers, and Facebook effectively used this data to show ads to similar potential customers.
For us, the key purchasing trigger is when someone moves to a new home, and we relied on Facebook’s signals in the pre-iOS 14 era to reach these prospective buyers. They would indicate their interest in buying a home or moving, and that’s when Facebook would show them our ads.
When you reach the third stage, it's essential to start investing in a scalable but slow-results channel. In our case, this scalable but slow-results channel was ATL (Above The Line) advertising. If you do not do this, at some point, you will hit saturation in the scalable paid media channels. So, invest in the scalable but slow-results channels.
We initiated ATL campaigns early on, focusing on markets like Kerala and Tamil Nadu, which are media-isolated markets. Because at some point, those campaigns started yielding results.
Once you reach a certain scale at a steady state, it will always be a mix. It will be a mix of your pure top-funnel campaigns, it will be a mix of content, and it will be a mix of three or four channels that work for you. This mix depends on the category, the stage of the brand, and the size of the brand.
Even today, when we launch a new category, such as our recent introduction of mixer grinders, we are still in the process of identifying unscalable channels.
The buyers of fans and mixers are quite distinct. While brand recognition and brand knowledge can be beneficial in some cases, the overall journey remains quite similar.
Attempting to scale channels right from day one, before determining product-market fit, often proves unsuccessful.
In this snippet, Arindam discusses Product Acquisition Channel fit and the critical factors that determine success with each channel.
There are primarily two factors to consider: cost and unit economics, and the intent and context of your brand within the chosen channel.
What is important for any of you to understand is how unit economics works for you. The Facebook CPMs are the Facebook CPMs. Say you’re selling a 200 product on Facebook. Your CPMs will be the same as that of someone selling a 10,000 product. You will not be able to make money in the first transaction.
It’s very important to have the right expectations because the CPMs will not change for you. The CTRs will also not drastically change for you. If the platform CTRs are 1%, maybe you can outcompete and you do 1.5 – 1.6% but more than that these metrics will remain the same for most brands.
Even when we started there were a significant number of searches happening on Google, but most of our keywords were being bid upon by Amazon, Flipkart, Snapdeal, Paytm. All of them had hundreds of millions of dollars to spend on marketing.
They were all operating at an LTV almost 100x that of Atomberg. Once you buy a fan, you don’t come back and buy fans again and again every month. For them, CAC of Rs 5000 was totally worth it. For us AOV was 3500 and there was no way we could afford that. This is why we decided we will not activate Google search. Maybe I’ll spend on Amazon Search and Flipkart search and that’s a much easier thing to do for us.
The other thing is the platform feeds the audience mindset and all of this.
Maybe for a lot of B2B products and all channels like Instagram and Facebook are not the right channels because people would not be in that frame of mind when they are coming across those when they’re browsing.
It is very important for brands early in their journey to figure out which channels work for them and which don't. Some of these calls are intuitive, but for most you will need to test the channel, understand the unit economics and customer mindset on it, and then decide whether to invest in it or not.
For example, you take the display as a channel. Display is a very on your face kind of a channel. But no one wants to look at display ads. You are browsing the open internet and you see those ads and you get a lot of cheap but low converting audiences for that. Display will always get those kinds of audience.
But say if you are selling a fantasy app like Jewel, some casino or something, something that people can start using by just clicking install, maybe for those you go ahead and do display ads.
If they were trying to acquire their audience, maybe for them display ads would work well, because they are appealing to a huge mass of people, and then you focus on whatever is the cheapest source of audience that you can get.