Today, in our blog, two teams will share their experience and insights on creating successful advertising creatives in product IT, teamwork, and app promotion on TikTok and Meta platforms.
Hello everyone! We are Diana Voloshchenko, Creative Team Lead, and Alona Horbunova, Performance Marketing Manager, and we work together at Universe Group. Thanks to our great collaboration, we’ve managed to bring the Cleaner Guru App from 35th place to TOP-1 in its category in just 3 months and have become a creative trendsetter in the niche.
Let's get acquainted and understand who is responsible for what in our team 🙂
The performance team is responsible for the app's marketing strategy, advertising budget allocation, and channel selection. The creative team translates this strategy into tactics, sets specific tasks for the production team, and brings them to life.
We have a saying in our team, "If the product works for 3 people, it will work for 1 million." Therefore, the marketing task is to ensure that 1 million users (or even more) use the app, not just 3 people. This is a driving force behind revenue growth and product development.
On the chart, you can see the dynamics of our app purchases. Each of these peaks represents a new creative or a new creative concept found.
Where to Find Creative Hypotheses
Let's move on to the process of making an ad creative. It consists of 6 stages, but we want to focus on three key ones: hypotheses, testing, and test results analysis.
Hypotheses must be based on market research. We look for them in the following directions:
Trends (it's important to consider whether they are suitable for your audience and are relevant to them).
Analysis of successful creatives in the market. We study trends at the idea, technique, and element levels and try to adapt them for ourselves.
Finding new tools. Any innovations can become a basis for creative or additional tools, same as AI avatars becoming popular now.
The product itself can be an important source of hypotheses. New features, mechanics in the product, audience behavior analysis — all these insights are the basis for making new creatives.
Competitor analysis. We carefully study common techniques and elements, trying to figure out what works there.
Audience Behavior Analysis. We research everything: deep metrics (VTR, comments, hook rate, funnel conversion), when users lose interest, and even read all the comments. The latter is a great source of insights. In our case, we immediately see what triggers people. Or it can be another direction—when people write about their biases against the product.
You can also create a dialogue with your audience through your creatives.
Here's what our process looks like 😁
Approaches to Creating Hypotheses vary depending on the set goals. We divide hypotheses into different approaches: conceptual, standard, and micro-testing.
Conceptual Approach (aka Concept Testing)
The basis of such a test is 75% of new concepts, something entirely new that we have never tested. Also, 25% of variations are included—creatives based on a well-known concept that we trust. For example, we have a hypothesis about a new mechanic—in the Cleaner Guru app, we show the delete function through a swipe. In the first stage, we demonstrate this function in different ways and then analyze which creative best conveys this mechanic to users.
Standard Approach
We use this approach most often. For instance, when we need to find a certain type or check if an idea works. Here, we reduce the number of creatives in a new concept to 40% and rely more on the known variations.
Based on these 3 creatives’ examples, we analyze why one creative worked better. Was it because the girl was a brunette? Was she charming? Did she resemble a celebrity? We develop a new vector through new hypotheses.
Micro-testing
This approach is used to find a specific detail or work on a specific idea. We create many variations to test their effectiveness.
If you find a top creative, the main task is to replicate it. One way to do this is through micro-testing. Besides increasing the chances of finding a top-performing creative, you also have the opportunity to find a new working element (testing new text, another color, shape of elements, etc.).
For example, we have found a working creative with a new intro, changing only the text or its appearance, which allowed us to find many more top performers.
So, all our test ideas and hypotheses led us to the best creative. This was our reward for a long path of experiments and analysis.
In conclusion: we have 3 approaches, each corresponding to a specific goal, helping us to achieve successful results in making advertising creatives.
How to Retain Your User and Even More
Of course, after making a creative, it's important for users to see it, and for us - to test it. Let's briefly explain our strategy using the funnel.
The goal of performance marketing is to attract users' attention and increase the number of purchases.
The largest number of users drops off at the Click on Ad stage. Therefore, our task is very logical: to make as many users as possible enter the funnel after clicking.
! It's important to analyze at which stage users drop off.
Sometimes we have creatives with a high CTR but a low conversion rate to purchases.
In our case, we have such a case with a very high CTR. The secret here is that the video resembling a carousel prompts users to click on the ad but does not clearly reflect the essence of the app, leading to a quick drop-off from the funnel. Instead, our top performer has a CTR of 10 times lower but shows better results.
Our testing setup rules:
The settings of the campaign for creative testing match the main purchase campaign settings. The only difference is the ad creative.
We limit the number of creatives in each ad set to 2-3, which are variations of one idea, to avoid confusion of the algorithms.
New concepts developed from scratch receive more test budgets and more chances for success as they help gain valuable insights.
Let's look at the main testing stages we adhere to.
Pre-test
The goal of this stage is to filter out the least successful creatives with a low budget using upper-funnel metrics like CPC or cost per install. Creatives that successfully pass this stage move on to the next.
We also have benchmarks for each funnel stage that help to use the test budget efficiently. Even a failed creative provides insights here.
Our goal is to find the best version of an idea and then proceed to the next testing stage with it. Even if all 10 ideas don't yield the desired result, we will find at least the best among them to refine and develop further.
Test
At this stage, we aim to find the creative that brings purchases at a satisfactory cost. We gather more data and statistics here as we allocate more test budget. This provides a more comprehensive analytical base for further optimization of creatives that need refinement: adding a hook, a background, etc.
We work with many countries worldwide, so if a creative doesn't work in, let’s say, the US, we can try launching it in other countries after adapting it to the local market.
Scale
Here, the best-performing idea is scaled up. We are satisfied when we have a successful creative that resonates with users. A nuance is to track how quickly a successful creative "burns out" and whether localized versions of the creative also succeed. We aim for our creatives to work in most countries, not just in specific regions.
Approaches to Scale Creatives
The task of a performance marketer is to launch as many new campaigns as possible with this creative to capture the market quickly. This can be through launching new campaigns or increasing budgets in existing setups.
From a creative perspective, quickly localizing creatives for all markets and creating their variations is crucial. Note: Localized creatives are a great tool if the product is also localized.
Results Analysis
Both teams analyze creatives but from different perspectives. A creative marketer looks for insights by comparing with other creatives: analyzing trends in ad sets, CTRs, and conversions, and trying to find trends for comparison.
The creative team analyzes detailed funnel metrics to understand at which stage users drop off and formulates new hypotheses to improve results.
When it comes to creative analysis from the performance team's side, the most crucial metric is the cost per action (CPA). They also carefully study the funnel, but their primary concern is to make the investments pay off, starting with the cost per user acquisition.
In the process, we've formulated important aspects of the interaction between the performance marketing team and creative marketers:
Close collaboration between teams. Everyone should know about any changes or new features in the product at the same time. The performance team can be sure that when new releases come in, the creative team is already working on it.
Understanding the tools of other teams and the ability to use them together to achieve goals.
Critical thinking about decisions made and readiness to discuss them with other teams to achieve the best results.
Quick communication among all participants allows us to solve tasks efficiently and quickly respond to changes.
And a cool feature: we have a bot that notifies us of successful creatives. This allows all teams to timely receive results and quickly react to them (prepare all variations in 2-3 days).
Our important lessons and conclusions
Numbers don't lie. We don't rely on subjective preferences when choosing creatives.
We believe that every "weird" idea has the potential to be successful if refined correctly.
Deep synergy with other teams is crucial for us, but we always stay focused on our goals.
Never give up: behind every successful creative, there are hundreds of failures. And that's not an exaggeration :)
Remember: there's no magic wand, but with a systematic approach, team speed, flexibility, testing, and analytics, everything will work out.
We hope our article has given you useful ideas and deepened your knowledge of the marketing field. If our approaches interest you, and you feel that this is your place in this industry, perhaps your professional success lies right here. We wish you successful cases and cool insights!
