From training data to how we ensure privacy, we're digging how Attentive AI was built to level up your marketing strategy.
This post is part of a new series exploring how Attentive is building AI and how marketers can use this technology to thrive in a new world.
Over the last decade, cloud and mobile technologies have had an undeniable impact on the way we work and live. They've introduced new ways of operating, sped up workflows, and have been widely adopted by engineers, businesses, and consumers alike.
AI has the same transformative potential. That’s why we’ve prioritized building tools tailor-made for marketers to help you stay ahead in an ever-evolving landscape, expand your team’s bandwidth, and create new opportunities to grow.
Attentive's AI tools, Automated Campaigns and Magic Message, empower marketers to streamline their workflows and grow their brands while maintaining the voice and personalized touches that set them apart. We’ve purpose-built personalization into our products to ensure that the copy and recommendations generated are unique to your brand, providing a customized experience for your audience.
With innovations happening at a rapid pace, it can be overwhelming to keep up with the latest tools and use cases in AI and machine learning. As someone who eats, sleeps, and breathes AI, I want to demystify these tools and make them feel more approachable (and less intimidating). Let’s dive into some of the most common questions I’ve heard and how you can think about incorporating AI into your workflow.
Check out The Marketer's Guide to Using AI—learn about the different types of AI, pros and cons, and how to incorporate it into your marketing strategy at different stages.
A quick primer on the evolution of machine learning models
In 2017, researchers in the field of machine learning applied a technique to enhance the performance of machine learning models by allowing them to allocate more attention to certain aspects of their inputs. This method, known as an attention mechanism, is loosely inspired by how the human brain processes information, where we can selectively focus on particular elements of our surroundings while disregarding others.
Attention has proven to be an effective tool for various applications in natural language processing, computer vision, and other areas of machine learning research. Researchers called this particular model a "transformer" because it effectively transforms inputs into outputs by honing in on the most important parts of the input to better inform the outputs. Since the release of the transformer model, enormous effort has gone into making these models larger and training them on more data. As a result, these models have become much more powerful and expressive.
One of the most amazing capabilities of transformers is that they can perform in-context learning. This means the model has been trained on a vast amount of data, giving it the power to solve new problems from just a few examples provided to it after training. However, it's important to note that while transformers have shown exceptional performance in various domains, their success is still limited by the quality and quantity of training data, as well as the complexity of the tasks they're designed to solve.
What data is Attentive AI™ trained on?
At Attentive, we take privacy seriously. That’s why we've developed a two-stage approach to building our AI products.
We start with a foundation of anonymized insights, which we use to identify trends, best practices, and patterns across our 8,000+ customer base. We then layer on your brand-specific data—including historical messages and performance data such as CVRs, CTRs, and opt-outs—to personalize the results to your brand and audience. This ensures that our AI tools are powerful and effective by being trained on a specialized dataset, while still protecting your privacy.
We use in-context learning by feeding in example messages to demonstrate your brand's voice to the model so that when it generates new messages, they sound like what you'd write. We take this a step further by using only a few of your best performing messages to make sure that the generated messages fit a similar profile.
How does Attentive AI come up with performant message copy?
Magic Message is designed to help you generate tailored, high-performing copy for your brand’s specific audiences and goals. To develop our AI-driven SMS copy assistant, we analyzed the anonymized message copy from millions of campaigns to identify the anatomy of successful messages.
We looked at patterns across high-performing messages, including:
- Campaign intent
- CTA formats
- Line breaks
- Message length
- Emoji usage
We use these insights to inform which copy suggestions to surface within Magic Message. But that’s just the foundation.
Is the copy Magic Message creates unique to my brand?
Each time you generate SMS copy with Magic Message, we blend in several of your high-performing messages to match your brand’s voice. This means Attentive AI can quickly learn to write in your voice after just a few uses—and only continues to improve as it learns to understand your campaigns. This approach allows us to provide effective copy suggestions that are tailor-made for your brand (and continue to learn your brand’s voice) while respecting your privacy.
What informs the recommendations Automated Campaigns makes?
We built Automated Campaigns to help marketers create and optimize multi-channel campaigns in a fraction of the time by leveraging the power of AI and data. Our AI analyzes anonymized insights from over 1.4 trillion data points, including:
- Sends (the volume of messages sent to subscribers)
- Campaigns (the type of message being sent)
- E-commerce data (sent to us by your integrated platforms)
- Attributable revenue
- Product information
We use these insights to identify which campaigns to include in a given cadence, such as which holidays show the highest lift, and ensure targeted messaging and precise timing.
In addition to this foundational data, we incorporate your brand's unique historical data to further personalize campaigns. For every event, such as a holiday or product promotion, Attentive AI determines each detail that goes into a successful campaign, including the right number of messages to include, the best send time to send, the optimal cadence, the target segment, and attention-grabbing copy that converts.
Automated Campaigns uses the same AI that powers Magic Message to create personalized and effective copy in your brand's voice. By using this approach, we ensure that our Automated Campaigns recommendations are tailored to your brand's individual needs, while building your data set responsibly.
How much control do I have over Automated Campaigns?
We know marketers want to maintain control over their marketing campaigns and add their own human touches, which is why we built Automated Campaigns with a high level of customization. You can edit the automated campaigns our AI generates, just like you do your manual campaigns. This means that you can customize the copy, imagery, and timing of the automated campaigns to ensure they feel bespoke to your brand.
We also send you alerts before sending out every campaign, so you’re never caught off guard or miss an opportunity to review and make changes. With this level of control baked into the product, you can feel confident that your campaigns are aligned with your brand and marketing goals.
At Attentive, we're committed to providing innovative tools that make it easier for marketers to harness the power of AI and automation. We want to help you become your team’s AI expert so you can confidently leverage these powerful tools to create more effective and efficient campaigns that drive growth and success for your brand.
To learn more about what you can do with Attentive AI, join the waitlist.