What even is an AI?

Ok, before we dive into our predictions it is important to define what we mean when we say “AI.” We have written an entire article explaining the artificial intelligence that underpins OpenAI’s ChatGPT and Google’s Bard, so this will be a very concise overview.

ChatGPT and Bard are both known as “Large Language Models” or LLMs, and are built on a digital architecture known as a neural network. Each of these models was trained on pre-existing (human generated) text which it then uses to determine what words are statistically most likely to follow one another. Using these statistical associations, LLMs are able to produce novel responses to questions that can easily pass for having been written by a competent human writer. The capacity to produce novel results instead of simply regurgitating predefined answers gives these systems their moniker “Generative AI” – and that is in fact what the ‘G’ of ChatGPT stands for. In addition to LLMs, there are AI models that operate under similar principles to generate digital art, edit photos, and even create music.

For the sake of this article, we are focusing on how LLMs, like ChatGPT and Bard, will be changing the landscape of small business marketing.

Prediction 1: The minimum viable level of marketing is going up

One of the most immediate impacts of generative AI is that everyone now has access to decent marketing. With a few hours of work, a person who has only a passing understanding of marketing can use AI to write reasonably convincing home pages, landers, and blogs. In the past this would have been outside of most business owner’s abilities, but now you can simply tell ChatGPT what you want and it will do the heavy lifting. This will mean that over the next few years the average quality of websites will increase as the lowest quality content is out-competed by AI-generated content.

Takeaway: If you want your company to remain competitive, your marketing efforts will need to stand out. A website that was good enough in 2022 is going to be lost in a sea of reasonably competent content in 2024.

Prediction 2: Big changes for informational search queries are inbound

Generative AI will change how users find information online. The proof of this is easy to see: both Google and Microsoft have invested heavily in the two largest generative AI models (Bard and ChatGPT) and are already rolling out AI-generated search result content.

Microsoft’s Bing already incorporates ChatGPT into its search results and Bard will almost certainly be coming to regular Google searches in the very near future. The search giants’ reasoning is easy to understand: why should users need to wade through several 1500+ word blog articles to find the small snippets of information that are relevant to their question?

Instead, generative AI gives consumers the ability to pose their question to Bard or ChatGPT and then be given a concise, personally tailored answer. And, to be fair, the death of these blogs won’t be much cried over. The vast majority of these informational articles are actually SEO pieces, heavily laden with keywords and surprisingly light on real content. Looking for a new recipe to try? Good luck! Need help figuring out a DIY project? Those top-ranked articles might take 1000 words to tell you that you need to call a professional, and then gently point you in the direction of their contact form. AI-powered search results will eliminate the chore of reading needlessly keyword-stuffed fluff and more quickly give people the information that they really want.

Takeaway: Informational queries undeniably represent a significant chunk of online search traffic and generative AI will probably mean that blog traffic will fall. However, we don’t anticipate these changes affecting your bottom line because of our next prediction.

Prediction 3: Searches for transactional keywords will remain unchanged

While people may happily allow search engines to provide a curated answer to questions like “Why is the sky blue” it is unlikely that consumers are ready to relinquish control of their purchasing decisions. People like choices, and are unlikely to happily accept an AI telling them which electrician they should call to install a ceiling fan or which plumber they need to contact to unclog their drain.

This isn’t just a hunch – as we currently see a huge portion of our client’s clicks are coming from Google Business Profile listings. When customers search on Google Maps they are presented with a list of businesses, complete with customer reviews, description of services, and pictures of their products, staff, or equipment. Customers have become quite savvy at using Google Maps results to find local service businesses that they want to work with. Currently, upwards of 33% of monthly incoming leads are generated from GBP listings where the customer never even visited the business’s website.
When customers need an additional level of certainty, they move from a business’s GBP listing to their website. We see that half of the clicks to our clients’ website come directly from their GBP page – not from traditional Google search results.

Takeaway: The importance of having a strong Google Business Profile is not changing any time soon. Customers will continue to use Google Maps to find local businesses and are not going to offload the decision making process to AIs.

Prediction 4: Data is going to become more valuable

Data has always been important, but the proliferation of AI is only going to increase its value. AI models require tremendous amounts of training data in order to learn and corporations have recently recognized this fact. When OpenAI was developing ChatGPT, data was cheap, bordering on free, but now CEOs realize the incredible amount of money that these systems can generate and they are pricing their data accordingly. The recent Reddit API pricing debacle is a great example of this.
Reddit’s CEO essentially came out and publicly said tech giants like Microsoft (owner of OpenAI) and Google were happy to pay the new, much higher, rates for API access in order to obtain the valuable AI training data that can be gleaned from Reddit’s communities. While this leaves smaller companies in a lurch, the simple reality is that data has become a more expensive commodity. The impact this will have on small businesses is yet to be seen – but we expect that AI companies will seek to gain access to business’s internal data to begin training the next generations of AI. By marrying conversion data with the website copy that produced it, future AI algorithms will be able to make more effective websites.

Takeaway: There are going to be inexpensive, or possibly free, AI services that will be able to process your company’s internal data and provide you with in-depth business analytics. The data that business owners provide will be used to train future AI models.

Prediction 5: The rise of AI chatbots

One obvious use case for generative AI is incorporating chatbots into websites, after all, this is essentially what LLMs do right now! The process for training the chatbot on your company’s internal data is yet to be worked out, but even if training starts out as time consuming or cumbersome, the process will become streamlined with time. These chatbots will be able to answer common customer questions, take payments, and even schedule appointments. It isn’t particularly far-fetched to imagine chatbots which will be able to answer phone calls as voice recognition systems and text to voice systems are already mature technologies.

Takeaway: Using a robust CRM and scheduling system which is able to integrate with AI chatbots will be important. Customers already express a strong preference for being able to schedule appointments online and are likely going to quickly embrace using chatbots for this purpose.

Prediction 6: Smarter smart homes

A promise that smart homes have made, and yet haven’t entirely delivered on, is that they can make life simpler. AI will bring this promise one step closer to fulfillment by allowing homes to communicate directly with service providers. At the very simplest level, household appliances that need routine servicing (hot water heaters, HVAC units, and the like) will be able to schedule their own maintenance appointments. Add in a bit of internal diagnostic capability, and appliances will be able to detect problems and then schedule their own repairs. As we mentioned before, customers are going to be unlikely to completely relinquish their control. However, consumers already sign up for annual maintenance agreements with service companies ranging from HVAC to pest control, and your smart home could simply be told in advance which companies it should contact should a problem arise.

Takeaway: Prioritize building long-term relationships with your customers. Service agreements are a great way to maintain customer loyalty and reduce customer churn. As mentioned earlier, having a robust CRM and scheduling system will put your company in a position to take advantage of these changes when they manifest.

Prediction 7: Ads will get personal

We’ve known for a long time that ads work best when they are curated to your specific audience – but at the end of the day we’ve been limited in what we can do with that information. Sure, you can define target demographics and create several different ads to match each, but at the end of the day they’re still targeting averages and not individuals. AI is going to radically change this paradigm, meaning that each customer will see ads tailored directly to them. This is not science fiction nor conjecture – Google announced that their Performance Max Google Ad campaigns will implement AI to deliver customized ads and even identify audiences that you may have overlooked. Instead of creating advertisements, business owners will instead provide Google with their core creative assets (videos, pictures, logos, and descriptions) which it will then use to create individualized advertisements.

Takeaway: Whether your business is already using PPC advertising or not, you should consider trying these new AI-powered campaigns and comparing their results with traditional marketing strategies. Additionally, building a robust portfolio of creative assets will help prepare your business to take advantage of these AI solutions – as the AI-generated ads will only be able to be as good as the inputs they’re initially given.

Screenshot from https://blog.google/products/ads-commerce/performance-max/

Prediction 8: Branding is going to matter even more

Company brand identity is important now, and will continue to remain important in the post-AI advertising world. In fact, we expect that branding will increase in importance as it becomes more essential than ever to stand out from the competition.
AI models do a great job at producing generic, compelling copy – but they cannot define your business. At the end of the day, having clearly defined value propositions and knowing why customers choose to do business with you over your competitors is essential. Be careful choosing value propositions which are based around nebulous claims like “customer service oriented” as every one of your competitors is already making that same claim, and with powerful language models at their disposal they’ll be doing so more eloquently than ever before. Instead, identify specific core strengths that separate you from the pack and ensure that your website and advertisements speak to that, and your staff and business practices embody that.

Takeaway: Having a clear sense of brand identity and narrowly defined value propositions will become more important. Vague promises of expertise or service will need strong justification in order to remain convincing.

Whether you’re ready or not, AI is going to change marketing

While the breathless promises and sensational headlines might make it feel that AI is all hype, it will cause major changes for small businesses. We can’t be certain of all of our predictions – but we are quite confident that at least some of them are all but guaranteed. Embracing these changes, or at least preparing for them now, may ultimately be a deciding factor between which businesses thrive and which struggle in the coming years. The good news is that these preparations will pay dividends immediately – a better website increases conversions today, a stronger brand identity boosts customer loyalty, and online scheduling makes it easier for your customers to work with you right now. In the long term, these small changes will position your business favorably to take advantage of the changes that AI is going to bring.

If you’re ready to begin making these changes, we specialize in outsourced marketing solutions for service businesses and can get your company ready for tomorrow.