How we generated last month's revenue ($8.7k) in 12 days with improved ROAS by 76%

Niche - Customised Gifting

The last 2 months before we achieved a 4X ROAS and 2X Daily revenue overnight were filled with a lot of testing and experiments, both on Ads and website front.

NOTHING COMES OVERNIGHT.

Tests and improvements on website - 

1. Reviewing Customer Journey – We added heat maps and viewed customer journey recordings and came up with action points to implement based on user-backed data. (over 5 major changes and numerous small changes implemented)

2. Changing the consumer journey flow - The client knew there were improvements needed on their website and was actively working towards changing things up before. The heat maps and recordings helped to get some much-needed perspective on the same to speed things up.

3. Introducing new products for Valentine's – they added 19 new products to cater specially for Valentine's, ensuring having products in all price ranges.

4. Separate flow for Valentine's – Separate landing pages catering to Valentine's special products were created, keeping the new customer journey in mind.

Tests & improvements on ads -

1. Narrowing down the target audience – Earlier the ads were being run on all placements, all genders and all age groups. Digging into analytics, we identified the age group with the most purchases and gender and correlated it with FB data to identify that one gender & age group was bringing a much better ROAS compared to the rest, and hence all new test campaigns were run on the same, ignoring the rest.

2. With age and gender defined, we also looked into product data. Instead of advertising all products, we focused on one product (highest ticket size), which was getting equal attention on ads but more orders. We developed an entire funnel for this one single product.

3. Testing placements – Instead of having one creative shown to everyone in every step of the customer journey, we asked the client to provide us with separate creatives for separate placements and separate creatives for different funnel steps (TOF, MOF, BOF)

Once the website was updated (all major changes) & ads were set for each funnel and each placement, all we had to do was to replicate the same learnings for new Valentine's products and pump in the money to scale things up.

After stopping all the experiments on ads, we immediately jumped to 3.5 ROAS, up from 2.25 during the peak of the experimental phase.

And once the ads for Valentine's were set, we hit, 

1. Highest grossing day ($1k) of sales since the start of the brand.

2. Best ROAS in the last 3 months (5.25) on the same day as the highest grossing day.

3. New winners for Valentine's (huge learning for future seasons)

4. And also identified 2 new winning products during the same time to reduce the drop in revenue post-Valentine's season and the discontinuation of Valentine's products!

All credit goes to the client and their team as well for being supportive of our efforts even with barely breaking even last month with all the experiments we went through, but the results were definitely visible this month!

We now plan to scale them up to $1.5k/day (Rs.1 Lakh) by mid-March and test and replicate the winning funnel to more product sets which we are yet to test!

Learnings - 

1. Test test test - Test on all levels. We tested placements, gender, age group, creatives, creative styles (video, photo, carousel, etc.), landing page, product page, description, and pretty much all aspects of the entire customer journey to crack the code.

2. Look at what data says – pretty much everything we did and finalised upon was backed by data. From video recordings to ad creatives to placements to creatives.

3. Understand your customer – The major learning was to understand what the customer wants and not what you want to show. One major website-level change was derived from this and changed from what the client had initially thought.

4. Being receptive to ideas – Pretty much all our ideas were appreciated, and freedom was given as well to test things out! 

Hope you find this case study useful and implement the learnings from the same!

Why are your ads not working? 

A comprehensive list of why most brands struggle with ads, created after reviewing, auditing and working with over 100+ client websites.

So, it’s going to be a really, really long post, but if the question above is the question on your mind, read it entirely and learn from the mistakes which others have made so you don't end up repeating them, OR take this as a guideline for things you need to take care of while working on your online store.

Bear with me, but I promise it will be worth your time to read this till the end!

6. New winners for Valentine's (huge learning for future seasons)

And also identified 2 new winning products during the same time to reduce the drop in revenue post-Valentine's season and discontinuation of Valentine's products!

All credit goes to the client and their team as well for being supportive of our efforts even with barely breaking even last month with all the experiments we went through, but the results were definitely visible this month!

We now plan to scale them up to $1.5k/day (Rs.1 Lakh) by March end and test and replicate the winning funnel to more product sets which we are yet to test!

Learnings - 

1. Test, test, test - Test on all levels. We tested placements, gender, age group, creatives, creative styles (video, photo, carousel, etc.), landing page, product page, description, and pretty much all aspects of the entire customer journey to crack the code.

2. Look at what data says – Pretty much everything we did and finalised was backed by data. From video recordings to ad creatives to placements to creatives.

3. Understand your customer – The major learning was to understand what the customer wants and not what you want to show. One major website-level change was derived from this and changed from what the client had initially thought.

4. Being receptive to ideas – Pretty much all our ideas were appreciated, and freedom was given as well to test things out! 

Hope you find this case study useful and implement the learnings from the same!