Why experiment with advanced mobile strategies?
You must commit to continuous experimentation in order to truly evolve and scale your marketing over time. Why? When customers interact with your brand through various mobile channels: text, in-app forms, push notifications, and WhatsApp, experimentation helps you move beyond assumptions about what resonates, with whom, on what channel, and when. Moreover, preferences and trends change over time, so you must test your content consistently to ensure you are evolving to meet your customers needs and interests.
Analyzing the conversion rates and revenue generated by each test will allow your brand to fine-tune your strategy so that customers receive the best path to purchase. This can include testing in the form of:
- A/B testing individual campaign messages with the omnichannel campaign builder.
- Comparing the performance of message paths or different messages in one path.
Review the core steps to A/B test mobile campaigns
Let’s refresh your memory on the core steps for running a successful A/B test for mobile-centric campaigns. Read through the steps below to ensure you know best practices at each stage.
If you create an omnichannel campaign, choose which messages within the path to test. If you choose a standard single-channel campaign, then that is the exact message you’ll be testing.
Start your campaign with a clearly defined, appropriately sized audience segment (e.g., recent mobile visitors who abandoned a cart, or new users who downloaded the app but haven't made a purchase).
Test 1 element of your strategy at a time, such as:
- The efficacy of emojis, images, GIFs, or video in increasing conversions.
- The ability for interactive prompt copy to spur engagement.
- Which buttons or linked pages yield more revenue.
- What mobile coupons or incentives spur the most clicks.
Note that you cannot test different channels within the same A/B test; for instance, you cannot test push vs. text. Instead, you could create separate pathways within an omnichannel campaign and compare results (we’ll show a few examples of this in the dropdowns below).
Build 2-4 content variations. We recommend testing no more than 4 at a time, as more can make it difficult to judge what drives significant impact. Have a control group in mind. They will receive either your current baseline mobile strategy or a variation with no intervention.
For instance, if you commonly send SMS without imagery, then SMS is your control. Meanwhile, your variations may include an MMS with a lifestyle-based image (someone happily using your product) and a product-focused MMS (a zoomed-in photo of the product).
Klaviyo will automatically share which variation outperforms the others once your test reaches statistical significance. This helps you easily pinpoint which mobile strategies significantly drive conversions. That said, the purpose of your test is not simply to identify a winner, but to act on insights. Replicate your high-performing content across future campaigns and avoid what doesn't resonate with your audience.
Identify elements to test in an advanced strategy
Wondering which elements are best to test across various channels? Let’s walk through what to test for each message component. Click the dropdowns to explore each, and remember to test only 1 element at a time.
Channel combinations
How to test: Compare pathways in the omnichannel campaign builder.
In the omnichannel campaign builder, you can create and compare multiple content pathways that share the same goal but use different channel types. This allows you to test which message types outperform others over time and in different order sequences.
Questions to answer during this time:
- Channel effectiveness: Which channel outperforms the others?
- Channel timing: When is the channel best used?
- Channel frequency: How many messages should be in your path?
- Omnichannel vs. single channel: What is the impact of weaving in multiple channels together rather than relying on one?
- Audience: For whom is the channel best suited (e.g., which segment)? Note that you can also find Channel Affinity insights on the profile level if your account qualifies for Predictive Analytics. Learn more about Channel Affinity.
Let’s go through an example. Say you want to determine which channel, email or text, outperforms the other when sent the day after a new product launch email.
- Path A: Product drop email day 1, Reminder email day 2, Last chance reminder text day 3
- Path B: Product drop email day 1, Reminder text message 2, Last chance email day 3
You can also use the behaviors of recipients to drive their journey. If you send an intro text message which they click but don’t convert through; follow up with a second text. For those who receive the text and don’t convert, you may follow up via email instead.
After the campaign sends, monitor the results of each within your omnichannel campaign analytics, focusing on which path drives the most engagement and revenue while avoiding higher unsubscribe rates.
Don’t worry: we’ll cover how to analyze and interpret omnichannel campaign data later on in the analytics section of this certificate!
Interactivity
How to test: Create an A/B test for either RCS or WhatsApp campaign messages.
Interactive elements in RCS and WhatsApp are meant to forge stronger customer relationships and drive action. Use an A/B test to find out what interactive components or prompts actually spur the most engagement among recipients.
For example, you might test 3 different call-to-action prompts:
- Variation A: You send a message asking the user to reply with a keyword to opt-in to a deal (e.g., "Reply YES for 10% off"). Very interactive and involves back-and-forth with subscribers.
- Variation B: You utilize an interactive message template featuring quick-reply buttons that users can tap directly (e.g., a button reading "Get 10% Off Now"). Somewhat interactive; involves a subscriber to click a button to get the deal.
- Variation C: You share a simple message with a 10% off coupon code and link. No interactivity involved beyond clicking a link.
Based on the results, you can optimize future messaging for maximum engagement and conversion via your most interactive channels.
Incentives
How to test: Create an A/B test for any mobile channel (text, WhatsApp, or push).
When A/B testing mobile-specific offers, your goal is to find the sweet spot between maximizing conversions and maintaining a healthy return on investment (ROI).
For a tiered discount strategy, test different offer thresholds to see what yields the highest profit and drives the greatest engagement. For example:
- Variation A: Offer "Free shipping for any order on the app."
- Variation B: Offer "10% off $50+ orders from our mobile app."
- Variation C: Offer "Get double the loyalty points when you shop on our app."
The perceived value of an offer often dictates its conversion rate, but the actual cost of the discount impacts ROI. Free shipping, for instance, can drive high conversions because it removes a common barrier to purchase, and its cost is fixed (or baked into product pricing), potentially yielding a better ROI than a larger percentage discount on high-ticket items.
Based on the results, prioritize a discount that achieves statistical significance in conversion uplift while improving profit margins.
Rich media
How to test: Create an A/B test for any mobile channel (text, WhatsApp, or push).
Media (e.g., images, GIFs, video, emojis, etc.) can draw the eye on mobile channels; however, in some cases (like MMS), they can increase your character count and cost of the send. That’s why it’s important to ensure each added design element actually drives tangible results that outweigh the cost or effort of sending it.
For example, you may create 2 variations of a text message for a major annual sale, perhaps your brand’s 20th birthday. You wonder: will an MMS with a birthday GIF draw the eye and encourage a click, or does it feel irrelevant to the user and even turn them away from engaging.
Your variations:
- Variation A: Sale launch with a brief but clear message to shop before it’s too late.
- Variation B: Sale launch with a brief but clear message to shop before it’s too late, showcasing a celebration-centric GIF.
Again, use the results to determine if future sales-related texts should feature images or GIFs, even if they’re not focused on 1 specific product.
Cost effectiveness
How to test: Monitor the effectiveness of messages in 1 omnichannel pathway.
A cascading cost strategy works by messaging your least expensive channels first. For those who engage but don’t convert, move them to the next most expensive channel, and so on until you finally reach social media ads, which typically cost the most.
This differs from how most brands operate, where each channel lives in a silo and profile data can't be shared across sites. By knowing that a profile has clicked an email and a text message, but still hasn't converted, you feel justified paying to serve them an ad on Instagram or via Google search. They are engaged, but they may just need one final nudge that an ad can support.
How this can help you: If you want to audit or improve your use of Marketing dollars and increase your overall return on investment. Take a step back and map out how much each channel costs, and begin to make wise decisions with each send based on cost per message.
For example, you may build an omnichannel pathway in the following order:
- Pathway:
- 1st send: Email to core engaged audience.
- 2nd send: SMS, RCS, or WhatsApp to those who open but don’t convert (based on channel preference and region).
- 3rd send: Targeted Instagram ad to those who interacted with both but never purchased (using custom audience targeting).