AI Chatbots Need Optimizers, Too

John Ebbert

To AEO
LLMs, CHATBOTS

On Saturday, WSJ reporter Katherine Blunt reviewed startup competition within the new world of “search engine optimization” for chatbots known alternatively as Answer Engine Optimization (AEO) or Generative Engine Optimization (GEO).

One startup profiled was newly-funded Athena:

“Athena’s software looks under the hood of different AI models to determine how each of them finds brand-related information. The software can track differences in the way the models talk about a given brand and recommend ways to optimize web content for AI.”

Chris Andrew of AEO tools startup Scrunch added blithely, “Your website doesn’t need to go away (…). But 90% of its human traffic will.”

Read more on the WSJ. (July 5 – subscription)

More:

  • “AthenaHQ – Get Your Brand Discovered on ChatGPT” (Winter 2025) – Y Combinator
  • “Scrunch AI launches out of beta to help more businesses win in the AI Search era” (March 2025) – Scrunch
LLMs, CHATBOTS

Podcast: No rush to optimize for chatbots

On a new episode of venture firm Greylock’s podcast, The Intelligent Marketer, CEO Ethan Smith of search and chatbot optimization agency Graphite said there’s no need for marketers to quickly rollout a chatbot strategy.

Mr. Smith advised waiting on AEO until the channel you are targeting is big enough -with one exception [begins 05:30]:

“There’s really no first mover advantage when it comes to AEO. It’s sort of a false concept. Whenever the channel is big enough, that’s when you invest in it.

If it’s not big enough right now, don’t invest in it…

For early stage companies, my first advice is, ‘Don’t spend any time on SEO because it’s not going to work.’ Do something else.

However, for Answer Engine Optimization (AEO), interestingly, it actually can work well for seed stage companies and the stuff that gets you from 0 to 1 is very different from the stuff that gets you 1 to 2.

1 to 2 would be: Meta, Google, SEO.

But, 0 to 1, can be these more unconventional channels and AEO actually can work for early stage companies because, again, to rank in SEO might take years. But as long as you’re getting mentioned by the citations for the question that you want to rank for in answer engines, you actually can rank pretty quickly.

So, depending on the stage of the company, I would decide whether to invest in it. But the first mover concept, I think, is generally not applicable.

So I just say invest in it whenever it’s big enough for you.”

Hear more on Apple Podcasts app (July 7 – 34 minutes).

Thought bubble I’ve said before:

The fact Greylock Partners – a very successful VC – is producing a newly-launched podcast on marketing shows that advertising and marketing technology has become an investable sector again in Silicon Valley. It’s reminiscent of the early 2010s when programmatic startups were receiving investments left and right.

The Valley sees that AI is going to massively transform advertising and marketing.

More: “Ultimate Guide to AEO (Answer Engine Optimization)” – Ethan Smith, CEO, Graphite

LLMs, CHATBOTS

Wall Street on OpenAI’s ad plans

Bank of America analyst Justin Post said: “If OpenAI’s ad-supported tier gains traction, there could be ad budget shift concerns for major incumbent platforms (primarily Google Search) affecting their growth outlook.”

IBD reporter Reinhardt Krause went deeper: “Google’s problem is that big brands and other advertisers could opt to move a big chunk of ad budgets to a new option — ChatGPT. For OpenAI, adding the digital infrastructure to sell advertising could be challenging. OpenAI would not want to damage the ChatGPT user experience.”

Read more on Investor Business Daily (July 7).

LLMs, CHATBOTS

Perplexity search ads

  1. “Perplexity AI’s Plan for Search Ads Looks Nothing Like Google” – quote: “Perplexity AI is slowly growing its ad business. Its main ad product is ‘sponsored follow-up questions…’” (July 7) – Business Insider (subscription)
  2. Perplexity has offered ads for half a year — marketers already want scale (May 28) – Digiday (subscription)
  3. Perplexity brings ads to its platform (Nov 2024) – TechCrunch
  4. Why we’re experimenting with advertising (Nov 2024) – Perplexity AI
SELL-SIDE

Podcast: Another AI idea for paying publishers

Annelies Jansen, chief business officer at ProRata discussed her company’s publisher compensation tools in the age of AI withBrian Morrissey, host of the Rebooting podcast.

She made clear that her company’s product line is meant for the 95% of publishers who don’t have deals with LLMs already in place. ProRata’s AI search solution includes Gist.ai, which is hosted on publisher websites and fed by content from member websites.

Ms. Jansen said, “We built Gist AI because we needed to prove that attribution is possible, and we needed to show publishers what it looks like [and how] we’re using their content to build an AI answer experience…”

This episode was recorded a few weeks before Cloudflare’s announcement last week about a new pay-per-crawl solution for charging LLMs.

Hear The Rebooting Show on the Apple Podcasts app.

More:

  • “ProRata Announces Gist.ai, New AI Search Engine Based Entirely on High-Quality Licensed Content” (Dec. 2024) – press release

BRANDS

Unilever designing AI-first

Unilever debuts AI-driven design unit to wean brands off TV-first model (July 7) – Marketing Dive

  • “The CPG giant developed the unit, called Sketch Pro, in partnership with IPG Studios, with the goal of delivering content three times faster for its brands.”

 

COMMERCE

Criteo & agentic commerce

Criteo chief product officer Todd Parsons reaffirmed his ad tech firm’s “vision for AI-assisted and agentic commerce” on the company blog yesterday.

Mr. Parsons used last week’s Cloudflare announcement – regarding a paywall for AI chatbot crawlers – as a pivot point for his company’s strategy:

“We believe the future of commerce will be shaped more by those who can empower AI responsibly, transparently, and inclusively, than by those who try to block it. And that means building connective tissue between AI platforms and the private, high-fidelity data that drives the sale of products. Let me be clear: This isn’t about ignoring the power of the content created by our thousands of publishing partners, many of whom were quoted in Cloudflare’s announcement. And this isn’t just about enabling AI. It IS about enabling the entire ecosystem — makers and marketers of products, retailers and marketplaces, content creators, and the agents themselves.”

Read about Criteo’s AI and agentic commerce positioning. (July 7)

More:

  • Video: “Criteo CEO Sees Agentic Shopping as the ‘New Website’ for Brands and Retailers” (June 2025) – Beet.TV at Cannes
  • Criteo CEO Komasinski at JPMorgan Conference: AI and Retail Media Focus (May 14) – Investing.com
AGENCIES

Quoting AI

“…So, the real advantage to this tech? Speed. Insight. And the ability to learn in days what used to take months. If you’re not exploring how to use AI to validate, iterate, and grow, you might be working harder than you need to.” – Horizon Media president Bob Lord on LinkedIn yesterday.

Mr. Lord joined the independent media agency in January after leadership roles at IBM, AOL and Razorfish. He reports into Horizon’s CEO and founder, Bill Koenigsberg.

SELL-SIDE

iHeartMedia’s contextual AI tool

IHeartMedia’s Sounder Is Using AI To Decode The Contextual Nuances Of Spanish-Language Media (July 2) – AdExchanger

SELL-SIDE

Podcast: about Google’s counterbalances

On the Risk Reversal Podcast, former TrueX co-founder and Fox executive Joe Marchese talks to host Dan Nathan about the implications of AI on the media and advertising industries.

Among Mr. Marchese “hot takes,” he emphasized the ongoing importance of “brand trust” in an AI world as well as Google having YouTube as a counterbalance to losing search traffic to LLM chatbots. He said:

“I don’t think you can put YouTube aside because YouTube is the ‘after Google.’ YouTube is the second largest place where people go to search for things. And if creators have more tools that use AI and make better shorts, films and the like… the time spent on YouTube only goes up.”

“(…)The other thing is (…) there is an innovator’s dilemma – a risk to this core [Google] business of search advertising, if search changes. However, counterbalance to that, Google is among the most likely to figure it out.“

“Now, once they figure out how AI best interacts with people, they have to figure out how to compensate publishers and how to get advertisers into that. And that is still an open question, so how quickly can you figure out a business model?”

Hear more on the Apple Podcasts app. (July 7 – 48 minutes)

PROMPT

In 5 years, how will the Llama large language model provide a differentiated advantage in advertising in comparison to other LLMs ?

Response from Meta.ai

In 5 years, the Llama large language model (LLM) is expected to provide a differentiated advantage in advertising through several key capabilities:

  • Enhanced personalization: Llama’s advanced understanding and generation capabilities will enable highly personalized ad content, tailored to individual preferences and behaviors.
  • Multimodal generation: Llama will likely support multimodal inputs and outputs, allowing for seamless integration of text, images, videos, and audio in ad creatives.

Read more on tipsheet.

AGENCIES

Agency compensation & AI

  • “How agencies are pricing AI—and what it means for industry compensation“ (July 7) – Ad Age (subscription)
  • “As AI Advances, Staff-Level Jobs at Ad Agencies Decline” (July 3) – Adweek
  • “How generative AI is Changing Agency Compensation Models“ (October 2024) – Read it on Medium, written by RAUS Global’s Christine Moore, former CFO at Dentsu and Mediabrands.
VENTURE

What is… Networked SaaS

“Why Networked SaaS is the new AI business model replacing per-seat pricing” (July 1) – Signal Fire, a venture capital firm.

Chris Scoggins, a former executive at marketing analytics firm Datalogix (sold to Oracle in 2014), is a partner at SignalFire.

MORE