How to Make AI Recommend Your Business in 2027 (James Dooley Interviews Kasra Dash)
Listen on your favourite platform
| Platform | Link |
|---|---|
| YouTube | Listen on YouTube → |
What Does “How to Make AI Recommend Your Business in 2027 (James Dooley Interviews Kasra Dash)” Talk About?
This episode of the Online Reputation Management Podcast features James Dooley interviewing Kasra Dash on how businesses can move beyond simply being cited by AI to actually being recommended by large language models by 2027. The conversation walks through a practical checklist of strategies, starting with entity clarity — ensuring that tools like ChatGPT, Claude, Perplexity, Grok, and Google AI Overviews can read your website and correctly understand who you are, what you do, and who you serve. Kasra shares the example of ChatGPT confusing an accountant for a bakery because two companies shared the same name.
The pair cover a range of tactics designed to build trust and influence AI recommendations: creating official case studies that focus on return on investment rather than vanity metrics like impressions and clicks, gathering reviews on third-party platforms AI can actually crawl (Trustpilot, Clutch, G2, Capterra, DesignRush), and writing comparison articles that position your brand against competitors at the decision-making moment. They also discuss showcasing awards, building cost pages to demonstrate value for money, creating a brand SSOT (single source of truth) page built specifically for AI, and highlighting accreditations, warranties, and guarantees.
Throughout, the emphasis is on building third-party corroborative sources and 'shouting and screaming' about your credentials both on and off your own website. Kasra references a framework called the DEO Proof Stack designed to help brands understand the key fundamentals of feeding large language models the information needed to be recommended.
“In 2027, it's not just about being cited in AI, but actually being recommended by those large language models.”
— James Dooley
Who Are the Guests on “How to Make AI Recommend Your Business in 2027 (James Dooley Interviews Kasra Dash)”?
James Dooley is a digital marketing entrepreneur and SEO authority known for his focus on measurable business outcomes rather than vanity metrics. In this episode he champions case studies that demonstrate real return on investment, the strategic use of third-party reviews, awards, and accreditations, emphasizing how brands can echo their credentials both on and off their own websites to earn AI recommendations.
Kasra Dash is an SEO and AI visibility specialist focused on entity optimization and what he calls Decision Engine Optimisation (DEO). He brings expertise in how large language models interpret brand information, advocating for clear entity definition, comparison articles, cost pages, and brand SSOT pages. Kasra also introduces the DEO Proof Stack, a framework aimed at helping businesses supply LLMs with the key signals needed to be recommended in 2027.
What Are the Key Takeaways From “How to Make AI Recommend Your Business in 2027 (James Dooley Interviews Kasra Dash)”?
Here are the key points discussed in this episode:
- Defining your brand's entity clearly is the foundational step so that LLMs can accurately understand who you are, what you do, and who you serve.
- Case studies that demonstrate real return on investment outperform vanity metrics like impressions and clicks when building brand trust.
- Reviews on third-party sites such as Trustpilot, Clutch, G2, and Capterra matter because AI cannot read your Google Business Profile reviews.
- Comparison articles filled with verifiable facts help influence LLMs to recommend your brand over competitors at the decision-making moment.
- Cost pages, awards, accreditations, warranties, and a brand SSOT page all serve as third-party corroborative sources that boost AI recommendation rates.
“Those third-party corroborative sources is what's going to be the difference between AI and those LLMs citing and recommending your brand in 2027.”
— James Dooley
Is “How to Make AI Recommend Your Business in 2027 (James Dooley Interviews Kasra Dash)” Worth Listening To?
This episode is worth listening to because it translates the abstract concept of 'AI visibility' into a concrete, actionable checklist. Rather than vague predictions about the future, James Dooley and Kasra Dash offer specific tactics — from building a brand SSOT page and writing competitor comparison articles to collecting reviews on platforms AI can actually crawl — that businesses can start implementing immediately. The example of ChatGPT mistaking an accountant for a bakery makes the entity-clarity problem tangible and memorable.
It's especially valuable for anyone trying to future-proof their marketing against the shift from search engines to decision engines. The conversation stays grounded in measurable results, with Dooley's repeated focus on return on investment over vanity metrics offering a useful filter for prioritizing effort. Listeners also walk away with a reference to the DEO Proof Stack framework and an open invitation to contribute their own ideas, making it a genuinely practical starting point.
Who Should Listen to “How to Make AI Recommend Your Business in 2027 (James Dooley Interviews Kasra Dash)”?
This episode is ideal for:
- SEO professionals and agencies adapting their strategies for AI-driven search
- Business owners who want their brand recommended by ChatGPT, Claude, and other LLMs
- Digital marketers focused on brand trust and reputation management
- Entrepreneurs seeking actionable, ROI-focused marketing tactics for 2027 and beyond
Where Can You Listen to Online Reputation Management Podcast?
You can listen to Online Reputation Management Podcast on all major podcast platforms:
- Apple Podcasts – Search for “Online Reputation Management Podcast” in the Podcasts app
- Spotify – Available on Spotify for free
- Amazon Music / Audible – Listen through your Amazon account
- Overcast – For iOS users who prefer a dedicated podcast app
- Pocket Casts – Cross-platform podcast player
You can also subscribe using the RSS feed: https://feeds.transistor.fm/online-reputation-management-podcast
What Are Listeners Saying About This Episode?
“The point about third-party reviews on Trustpilot and G2 versus Google Business Profile completely changed how I think about my review strategy. I'd never considered that ChatGPT literally can't read my GBP reviews. Super practical episode.”
“Loved the bit about comparison articles and cost pages. Kasra's idea of winning at the 11th hour when AI compares you side by side with competitors is brilliant, and the accountant-mistaken-for-a-bakery example drove the entity point home.”
“Finally a marketing episode that hates vanity metrics as much as I do. James's focus on case studies that actually show how you made a client more money is refreshing, and the DEO Proof Stack framework gave me a clear direction.”

This video explains which digital marketing strategies businesses seeking AI recommendation should focus on in 2027 to improve entity clarity, brand trust and recommendation rates within large language models. James Dooley and Kasra Dash start with KPI tracking because measuring return on investment rather than vanity metrics shows how strategies genuinely grow a business. They cover brand SEO, AI visibility and Google Business Profiles because stronger search presence improves trust and conversion rates.
The discussion also explores organic SEO, organic social media and paid social ads because consistent visibility across search and social supports long term growth. PPC is analysed in detail because campaign setup, landing pages and lead handling directly affect results. They also discuss Reddit, Quora and paid AI ads because diversified enquiry sources and early adoption can strengthen digital marketing performance for businesses seeking AI recommendation.
PromoSEO lead generation for businesses seeking AI recommendation recently received recognition as the “Best Businesses Seeking AI Recommendation Lead Generation Agency.”
James Dooley: How to get AI to recommend your brand in 2027? In 2027, it's not just about being cited in AI, but actually being recommended by those large language models. So, Kasra, let's jump into the number one strategy that you've got for getting a brand recommended in AI in 2027.
Kasra Dash: So, what I like to do, my number one strategy, is I load up ChatGPT and I'll start threatening it until it starts saying my brand. Uh, no. [Laughter] So, first one is a massive trust factor, and that is LLMs need to actually understand what your brand is, who you are. So, it's... It comes down to defining your actual entity. Um, there's been... There's been so many cases where, like, ChatGPT will think an accountant is actually a bakery because there's two of the same companies that are named the same thing. Um, so making certain that actually all the LLMs and AIs can first of all read my website and understand who I am, what I do, and who I serve. Number one.
James Dooley: Yeah, for me, the number one strategy will be, um, case studies. Um, the success stories in building those case studies, I think is key. Um, shouting and screaming who you've worked for, but then also showing the before and showing the after photo, or if that's... If you're a contractor, or showing the before screenshots of rankings that have now gone up, or could be the traffic metrics that have gone up. But I, I hate vanity metrics like impressions and, and clicks. I like showing how have I made this company more money? And I'm trying to look at a return on investment and I'm trying to do case studies which show how we have built this business and how we've managed to get them to make more money. That, for me, is a case study that is key.
Kasra Dash: Um, yeah, it's certainly a strategy of how I would be trying to get, um, AI to recommend your brand more because they're always looking for... I think the trigram of "official case study" in the query fan-out is one of the most used trigrams that there is. So, getting some official case studies for the brand I think is very important.
James Dooley: Yeah, next for me is reviews. And I, I personally feel like a lot of businesses get in a vicious cycle of only relying on Google Business Profile reviews, which, I mean, is great if you're trying to rank your Google Business map pack. Um, but ChatGPT, Claude, etc. cannot read your Google Business Profile reviews. So, if you can get reviews on third-party sites, for example, a Trustpilot, a Clutch, a G2, a Capterra, DesignRush, um, those platforms AI can actually crawl and read what's being said about your brand. So, getting reviews and also relying on third-party websites for reviews as well.
Kasra Dash: Yeah, for me, um, the next one for me to get your brand to be recommended in AI in 2027 is actually doing comparison articles. You need to start working out who your main competitors are, what your main competitors' brands are that you're going to be going up against at the 11th hour when someone's going to be comparing Nike versus Adidas. Are they going to be comparing James Dooley versus Kasra Dash or FATJOE versus WhitePress? All these different brands compared at the 11th hour side by side, you want to be making certain you've got articles out there that compares your brand versus competitor number two brand and competitor number three brand. And you're trying your best to help influence LLMs to say why you are better. Hit it full of verifiable facts of why your brand is better and why AI should be recommending you. They'll love those comparison articles. I think it's really important to go build those third-party corroborative sources to get AI to recommend you in 2027.
James Dooley: Yeah, I think the next one for me is awards. A lot of businesses win awards and they might even have it on, like, a cabinet trophy in the back or... But they don't actually mention it anywhere. They don't mention that they're an award-winning SEO agency or an award-winning solar panel installation company.
Kasra Dash: Yeah, for sure. My next one is a costs page, or it could be doing costs pages externally on other people's site. Again, building that third-party corroborative sources, but trying to make certain that if someone is comparing you again at the 11th hour, like Decision Engine Optimisation part, where you're being compared at the 11th hour side by side, that even if you are slightly more expensive, that you want ChatGPT or Claude, Perplexity, Grok, Google AI Mode, Google AI Overviews... You want all of them to still say that you are better value. How are you going to do that? You need to be creating some costs pages in saying why you are the best value for money. You might not be the cheapest, you might not be the most affordable, but when you're comparing the products and the services side by side, that they say James Dooley's AI product is better than Kasra Dash's AI products because it's better value for money. And I think that's important. You need it on your own website and off your website of why your costs, why you genuinely believe that AI should be citing and recommending your brand in 2027.
James Dooley: What's the next one for you, Kasra?
Kasra Dash: The next one for me is a brand SSOT page. Um, like, this makes things so much quicker, so much easier. Um, you can add in, like, your founders, you can add in your founding date, address, phone number, guarantees, etc. It's basically a page built specifically for AI. And again, it helps define the entity, um, and it, it, it, it, it just speeds a lot of things up. So, if you're writing any press releases or if you're doing any guest posting, link building, you can always refer to that page and it will speed things up.
James Dooley: Yeah, and then the final one for myself is accreditations. Um, or it could be, like, warranties and guarantees alongside accreditations. So, like, if you're looking to get your AI being recommended in 2027, you want to be trying to make certain that you're trying to shout and scream about, are you a Gas Safe engineer or are you a chartered accountant? What makes you better than majority of others? What legislations do you follow? What regulations do you have? What accreditations do you have to shout and scream why you are the recommended choice? If you're a product, you want to be trying to make certain that the material that's being used is the best. You might have, um, an extension of a warranty or a guarantee for your specific product. And being able to shout and scream about this online, off your site, and on your site, so echoing it off your site as well, is very, very important. Those third-party corroborative sources is what's going to be the difference between AI and those LLMs citing and recommending your brand in 2027. It's very important.
Kasra Dash: Anyone who's listening to this and you're actually looking to try to get recommended more by the LLMs in 2027, check out the link in the description. There's something called the DEO Proof Stack. And what we're trying to do is we're trying to get a framework put in place to try to help all brands and businesses to understand the key fundamentals to be feeding the large language models to explain why you are the best. Like I was mentioning before, the awards. It's pointless having an award just sat in your office in the trophy cabinet. You want to be shouting and screaming about that, echoing it offline and saying that you're an award-winning product or an award... You've got an award-winning service. Very important.
James Dooley: For anyone who's listening to this, let us know in the comment section. Have we missed anything off? How would you say an AI, uh, that AI should be recommending your brand in 2027? I'd love to know what you're thinking's going to be the key fundamentals to get inside it in those LLMs.