These days, it seems like there’s an AI tool for everything. Now there’s even an AI tool to manage your AI tools.
On Thursday, artificial intelligence startup Anyword released what it calls a “gen-AI performance platform” that analyzes and scores all the content generated by the various tools in a marketer’s stack.
“We believe you’re going to use multiple AI applications and platforms, not just one, and we want to help you do that,” CEO and co-founder Yaniv Makover told AdExchanger.
The platform integrates CRM systems and social advertising channels, as well as AI models, such as ChatGPT, and is able to advise marketers on the type of data each model will need to deliver the best results.
Custom content
The company was launched in 2021 (just before the generative AI boom, Makover noted) from a machine learning-focused company called Keywee.
With Keywee’s suite of tools, media companies like the New York Times and NBC can analyze their news content to predict audience reaction, but not to influence the newsroom itself. The technology didn’t exist at the time.
Anyword was designed to address this need by ingesting all of a marketer’s publicly available content (ad accounts, social media pages, websites, email campaigns, etc.) and then making suggestions on how to tailor the new content to a specific audience.
The platform achieves this by using Retrieval Augmented Generation (RAG), which enhances the output of an AI model without requiring elaborate prompts from the end user by retrieving additional information and context from a large database before generating content.
Anyword’s RAG was trained partly on a huge test dataset the company already had and partly from publicly available performance data (although no copyrighted content, Makover pointed out).
Data from users of its free tier is used to improve the model, but not data from enterprise customers like Vimeo, Dollar Shave Club or AlphaSense, which are siloed in their own private versions.
So far, the company claims to have generated a 20% increase in sales for B2B, B2C and enterprise marketing teams using the platform. Customers can also track different metrics based on their broader business goals, such as cost per acquisition or click-through rate.
Art versus science
But Anyword’s main selling point is that its platform will be able to help marketers achieve a consistent and easily controllable brand tone across channels, especially across different AI models.
Of course, a full-time writer can’t help but wonder: if so many AI tools and platforms are needed to ensure a consistent brand voice, wouldn’t it be easier and less expensive to hire a writer with a background in brand marketing?
Makover had an answer for that.
An AI model can’t outperform a marketing professional, he said. But what is the tool? can potentially do better than humans is to support its copy suggestions with quantitative data.
To test this hypothesis, in 2019, the Keywee team gave content that had already been tested for a particular target audience to a number of copywriters and asked them to predict which one would perform best.
Apparently, these writers could only predict the best-performing text 60% of the time, frequently disagreed with each other, and even contradicted themselves when shown the same text three months later.
Even so, Makover acknowledged that a marketer using the Anyword platform might have a reason to choose a less-performing text variation. The important thing is that marketers have the metrics they need to make informed choices rather than relying solely on gut instinct.
Co-founder Omer Rabin, whose appointment as Anyword’s chief revenue officer was also announced Thursday, recalled a similar conversation he had with a client of the brand about Anyword’s tools.
“None of you came here to be marketers so you can focus on the science,” one brand executive reportedly told employees. “By bringing a solution like this, we’re covering the science so you can focus on the art.”
Part of the workflow
The overall role of AI in the everyday life of the average worker presents a similar contradiction.
Although most tools are designed to save time, a recent Upwork Survey suggests that most employees feel the exact opposite: AI tools have actually increased their workload.
Asked about this, Makover did not seem surprised.
“I think there’s a lot of pressure from management teams to embrace AI, and it comes at the expense of using best-in-class tools,” he said, noting that it’s difficult to sort out platforms that rely on generative AI to operate, like Anyword, and those that are simply trying to capitalize on the hype around AI.
Still, he remains optimistic about AI adoption and learning. “The applications are getting better and more useful, and I think that’s going to continue,” Makover said.
