With the launch of ChatGPT in November 2022, AI entered the public sphere. Natural language processors like GPT have made AI accessible to anyone with a keyboard and a grasp of grammar, leading to an explosion in hype that has put artificial intelligence on center stage ever since.
Two years on, AI is moving from hyped potential to practical application. Yet AI integration doesn't look like science fiction for businesses -- its most powerful impact is behind the scenes. Real-time data analysis and its immense predictive power can play a significant role in your startup and the customer or user journey. AI may be making waves but this technology is still in its infancy, so founders and entrepreneurs should prepare to be surprised by AI's uses and impact across their business, and ready themselves to leverage it wherever they can. Here are some of the core changes we've made using AI at Atom - all adding up to a better experience for our users.
Search and discovery should put the right products in front of the right customers, and AI can determine both. Firstly, deep categorization of product criteria enables better classification. Secondly, it matches these better-understood search results to individual user profiles.In our domain marketplace, we've trained our AI on a huge genome of root words, synonyms, and metaphor as well as a range of domain characteristics (including name style, length, extension, and number of syllables). By understanding these characteristics and matching them to specific use cases, we provide our users with the best possible results when they search. For example, PurityCompass.com would be matched to generic search terms that buyers use such as 'beauty' or 'fashion' or brand positioning keywords like 'elegant' or 'natural'.
Our AI also employs individual user personas to critique and improve its own results, ranking domains linked to a searched term based on how likely they are to appeal to the user. The AI builds these detailed user personas based on the domain names people click through, how long they spend on a landing page, what type of names they return to. These variables are used alongside search terms to build deeper insight into the brand our customer is building and customize search appropriately, essentially creating personalized and ever-improving results.
GenAI is capable of producing compelling descriptions as well as visually rich imagery to promote products and services. We have found recently, however, that GenAI image creation can be used to create an additional product that adds value for the buyer.
A few weeks ago, we began testing integrated AI-generated image creation for our domain sellers. Our tool allows them to create branded imagery for their domains including logos, merchandise, and virtual billboard creation. The billboards have been a particular success, turning an abstract domain into a tangible brand. These tools are proving popular - within 3 weeks of launch, sellers created 85,000+ images. These images assist our sellers in marketing their domains, but they have also improved buyer experience. Multiple buyers have requested downloadable versions of these visual assets for presentations and pitches, justifying domain purchases to their board or C-suite.
Clearly, what began life as a marketing tool -- appealing imagery highlighting a domain's use case to brands -- has become a valuable asset to the buyer as well as the seller. The use cases it provided are interesting as they were a surprise to us! We knew the billboards and logos would improve sell-through and add a unique touch to landing pages, but we hadn't envisioned their role in supporting internal decision-making with end users. That's one of the great things about new technology - it may do what it says on the tin, and a whole lot more besides.
AI has accelerated marketers' ability to target the appropriate customers and personalize the content they see. By analyzing user data and behavior at scale, AI builds detailed audience segments and puts your ad campaigns in front of the most relevant high-intent users.
This generates meaningful awareness in the right demographics and positively impacts ROI and conversion rates. We recently implemented Google's AI-powered ad solution Demand Gen, which targets relevant users across YouTube and Google Search. With this focused targeting, we saw a 57% more efficient cost per acquisition.
AI can also leverage user data to provide enhanced lead scoring, enabling your sales team to put effort in where it's most likely to be rewarded. With traditional lead generation, around half of prospects are likely to be a poor fit for your products or services. Integrating a well-trained AI into your lead scoring hones your efforts and, even better, it iterates and learns with every round.
Some products or services can't come with a 30-day test period or a money-back guarantee. In these cases, getting the customer over the line and purchasing can be the toughest part of the sales funnel. AI, of course, can help.
We designed an AI Brand Alignment tool to help our buyers get an objective opinion about how well a specific domain aligns with - and will therefore support - their brand vision. This allows users to determine how well the brand names they are interested in fit their specific goals.
Tools like this can be useful when buyers aren't completely sure about a purchase. Particularly with larger purchases the advice of a "virtual consultant" can reassure them they're making a good choice both for now and looking to the future, nudging them from interested to purchasing. The AI Brand Alignment tool also offers suggestions that may be more suitable for a brand if there's not a match between the domain a user is interested in and their brand vision, so it's never a dead end with the potential to lose customers at the last hurdle.
Dynamic pricing isn't new: the most basic economic principles determine that prices have always fluctuated based on demand. AI, however, enables real-time dynamic pricing that optimizes revenue for ecommerce brands. AI is now capable of analyzing thousands of data points, including market demand, customer behavior and external factors including political events and social media trends. The result is exceptionally flexible pricing models, such as that which allows Amazon prices to change every ten minutes. Pricing has never been this dynamic, and we're still grappling with the ethical implications. Personalized pricing, sometimes less generously called surveillance pricing, boosts profits by 19% relative to optimized uniform pricing, but the Federal Trade Commission (FTC) recently launched an investigation into the practices of some companies, including Mastercard and Chase.
Despite some controversy, dynamic pricing based on sophisticated AI algorithms can be a win-win for consumers and brands, as it improves inventory management and (when used well) leads to fairer prices based on economic principles.
From demand generation to personalized pricing, the ability of artificial intelligence to gather and analyze data leads to its most powerful uses. AI can also monitor sources of information such as social media, and interpret this data to build a complete picture of what consumers are saying about your brand.
This includes metrics such as brand awareness and aggregated consumer sentiment. Importantly, the data that AI gathers isn't simply quantitative, but qualitative too: it can identify, for example, why customers choose you over a competitor or vice versa.
Cross-channel sentiment analysis with AI gives you real-time data and deeper insights into brand performance, allowing you to make accurate branding and marketing decisions.
As powerful as AI seems today, its impact on business is in its infancy. I recommend paying attention to beta releases of AI tools as well as keeping an eye on how your competitors are implementing it so you stay current on the applications and limitations of AI.
AI will only become more powerful in the coming months and years, but that doesn't mean you should wait to begin using it where it makes sense. Startups should begin building their AI infrastructure now to be ready to leverage each and every relevant technological advance. Working with AI now not only means you'll understand how to use what works, but also that you'll know what to ignore because it doesn't fit your needs. At this point in the growth of AI, exploring how it can be applied well to your business model is key.