9 Use Cases HyperGrowth Partners Use AI - pt. I
Creative LLM Use Cases to Drive Growth Across SEO, Outbound, and Performance Marketing
Rapid advancements in AI are revolutionizing growth strategies across almost every domain — product, marketing, sales, and design.
This transformative period is taking shape in two main directions:
Horizontal AI. Leveraging non-specialized Large Language Models (LLMs) like ChatGPT, Claude, and Perplexity to streamline general information discovery, gathering, and organization workflows that would have taken a lot more time to complete without it. This tech has such a broad set of use cases that it’s quickly shifting across the adoption curve — from early adopters to the early majority — in just over a year.
Vertical AI. Utilizing innovative solutions from specialized AI startups to inject automation and intelligence to reinvent very specific use cases, like lead list building, copy personalization, and more. Compared to the horizontal AI market — which is fairly concentrated — there is a plethora of vertical AI apps that are emerging across the whole domain spectrum. This makes it very hard to distinguish the good from the bad and the ugly. However, because these use cases are more niche, they’ll take more time to get popular; hence they represent a greater opportunity for a competitive moat.
Moreover, AI evolves at an exponential pace; whereas our human brain does so linearly. This discrepancy is creating a significant bias around the adoption of these AI tools, especially vertical ones. Experiments, that may fail today, could potentially succeed in a few months. For example, AI voice trying to replace SDRs might not work now, but it could work just fine by Q4 2024.
Unfortunately, most organizations do not revisit past experiments with sufficient frequency, leading to a linear progression of learning and biases in contrast to the exponential evolution of AI tech. Thus, we’ll need to factor in and distribute the cost of testing across different orgs.
To sum it all up, sole reliance on generalist AI like ChatGPT is already an industry norm in tech, foregoing any competitive edge. On the other hand, sifting through the plethora of emerging vertical AI apps requires expertise to identify truly valuable tools amidst a sea of mediocrity.
How can growth teams employ AI creatively to enhance growth and realize cost efficiencies in their daily operations? In this three-part series, we’ll introduce nine practical use cases from HyperGrowth Partners across three main growth domains: SEO, outbound, and performance marketing.
on AI-Assisted SEO & Product Marketing
SEO has been incredibly sensitive to the rise of LLM, both for technical and content creation purposes. However, most people use AI in SEO for keyword scraping, analysis, and content creation; but we’ll see that teams can get creative with exciting use cases across content and product marketing. Let’s see some of them.
1. Use Writesonic to Create High-Quality Content with AI
As AI dramatically lowers the barriers to creating content, there’s a reason to believe most AI-generated content is spammy; but the truth is that it doesn’t have to be. AI can also create informative and educational content that doesn’t just rank, but it’s also high quality.
Despite there being a handful of tools for AI content creation, Writesonic seems to be among the best positioned for the creation of high-quality, technical, long-form content. In this context, technical content is defined as topics that are fairly niche and for which there’s no log of content on the Internet. Hence, horizontal LLMs don’t have content to scrape to inform the generation of fresh content.
To fill this gap, Writesonic lets you upload or link reference articles, and rich tone of voice docs to ground their model with defined and technical guidelines that increase the output relevance.
For example, we used Writesonic to create a 10,000+ word technical guide on real-time analytics with one of our clients by feeding the LLM with subject matter expertise from previous work, effectively saving countless hours of work to create the same output.
This content doesn’t just rank, but it also generates product signups at scale; which is outstanding considering the amount of time invested in the creation and editing of this content.
However, it can’t be stressed enough that — despite saving all those hours from content creation and proofreading, human resources must be allocated to edit AI-generated content to minimize hallucinations.
Why? Not because, as most people think, Google will penalize you for AI-generated content per se. Google doesn’t do that, or care about that too much. What Google cares about is maintaining a good user experience, as measured by clicks, CTR, and engaged sessions. That’s why, if your users make the effort to search for your keywords, click through your article, but then see something weird when reading it, they’ll disengage. This is what ultimately hurts not just your rankings, but most importantly, your brand credibility.
In the future, fact-checking APIs like Aspire will be able to help fact-check AI-generated content across large data sets to help minimize these hallucinations at scale; but a closed-loop workflow that solves this problem doesn’t exist yet.
2. Use Humata.ai for Qualitative Market Research at Scale
This use case is especially relevant for Enterprise sales, where the sales organization has tens of customer calls every day; with transcripts stored in sales intelligence tools like Gont, Grain, Fireflies, or tldv.
Humata.ai is an AI chat UI — similar to ChatGPT — that specializes in parsing information across long documents. You can upload a large number of PDFs to let Humata extract valuable insights at scale — like keywords, blog title ideas, and other content plays — which can then be handed over to Writesonic or other tools for creation and articulation.
How does it work in practice?
Upload your PDF with transcripts from customer calls, community comments, etc.
Ask questions in natural language to extract customer pains, solutions, and patterns.
Target each pain point or pattern by creating:
Specific SEO content addressing them.
Inform messaging & positioning,
Personalize outbound emails
Humata is a great example of a horizontal AI app that can be verticalized depending on who is using it. You can get creative and even use it to streamline the onboarding of new hires and provide them with always up-to-date information about the customers, the company, and the market.
Best of all, Humata has private data rooms so your PII and customer data are safe; which might not be the case with other horizontal AI startups.
Need help injecting AI in your growth workflows from experts who have done this multiple times at scale?
According to Kevin Indig: “If you haven’t used this already, ChatGPT Chrome Extension is a must-have in your growth toolkit”.
He’s used the extension with HyperGrowth Partners to create data-driven programmatic SEO campaigns that scrape the publicly available data to create programmatic content at scale.
From our dedicated guide on the topic: “Programmatic SEO means building thousands of web pages that target specific long-tail keywords. Relevant examples include:
Tripadvisor has thousands of pages ranking for ‘best {thing to do} in {city}’
TimeOut for ‘best {restaurant} in {area}’
Glassdoor ranks for ‘{job title} salary’
Wise has tons of pages ranking for ‘best {exchange rates}’
Statista ranks for thousands of pages for ‘{category} - statistics & facts’
Causal has recently ranked for ‘calculate {financial formula}’
Thomas Cook ranks for thousands of ‘{location} weather in {month of the year}’
These are all examples of programmatic SEO or pSEO for short.”
pSEO relies heavily on large data sets to scale. And what better way of using AI to scrape, cleanse, and summarize this data effectively? Even better, when data is publicly available through API, it’s always up-to-date and fact-checked; which helps greatly in minimizing hallucinations.
So, how does the process work in practice?
Design a prompt for the extension to retrieve data and fill in your pSEO search query. For example, if your pSEO query is: “Average salary in {us-state}”, {us-state} is your key variable of a data set for each of the 50 States in the US.
Run the prompt to scrape data automatically from public APIs, and re-run the prompt regularly to update your output.
Map the data from Sheets into your CMS with Zapier or custom scripts to land all the data points and AI-generated summaries into your landing pages.
As you might have guessed, a big part of this process is to ensure you have the right prompt to avoid common patterns LLM falls into and avoid content being bland and non-informative. So, ensure you have a good prompt and see your organic clicks scale.
Get the next 6 AI use cases for Outbound and Paid Marketing
It’s no news AI is the revolutionary tech of our decade.
In this post, we’ve seen the pros and cons between horizontal and vertical AI tools and showed real-life use cases of how you can use AI to create technical content at scale, run programmatic SEO campaigns, and even parse your customer call transcripts to extract insights for multiple use cases.
In parts II and III of this post, we’ll dive into more AI use cases, specifically for Outbound and Paid Marketing.
 | A guest post by
|