In our SEO agency, we onboarded a SaaS client who wants to do SEO just to rank in LLM-generated responses.
This is the first time we’ve seen requirements like this in our agency.
The reason could be the increasing popularity of generative AI tools. Here’s the global search demand popularity of ‘ChatGPT’ (last 5 years):
After extensive research and implementation, we have created a process that any brand can follow to increase its brand presence on LLMs.
This process has significantly increased brand mentions in LLMs.
I have created a checklist out of this guide using ChatGPT. Download it and start implementing it for LLM optimization.
👉 Download LLM optimization checklist
Step 1. Audit the current state
First, create a list of LLM chatbots (ChatGPT, Claude, Perplexity, Gemini, etc.) where you want to increase your brand awareness— because not all LLMs work on the same principles.
Next, see whether the chatbot mentions your brand name in the ‘top X tool/service’ type of search terms.
For this, you can try different types of search terms in LLMs
What is {product_name}?
What are the top property management software?
The goal here should be to understand:
FIRST: What information does LLMs have about your brand?
SECOND: Whether these LLM chatbots recommend your brand in the best product category or not?
Here’s an example of how you can document the answers:
Next, create a list of the most recommended brands in your niche/industry and their citations (from where LLM chatbots are finding the information):
This will give you clarity of where you’re standing right now compared to the competitors.
TIP: Also, notice the related questions shared by Perplexity about your brand name. This way, you can understand how LLMs associate your brand with different entities.
Step 2. Analyze the source types used by chatbots
After we analyzed all the top sources, we found various types of sources from where these LLMs are finding the information. This includes →
On-site website
YouTube videos
Social media platforms (specially, LinkedIn bio, marketplace/directories, Instagram shorts, Reddit, Wikipedia, listicle articles, About/FAQ pages)
If the brand is pretty new, start with creating foundational backlinks and directory submissions.
Ensure these third-party websites get indexed in Google— otherwise, there is no point.
Why is this important?
Knowing the exact source type and even exact page URLs can help you replicate the same for your brand.
(We will talk about the same later in this post).
👉 I highly recommend reading my newsletter to understand how Google associate your brand with different entities.
Step 3. Start with the basics of LLM optimization
Brands often neglect the power of optimizing webpages that they have full control such as:
Website homepage
Product landing pages
Branded FAQ page
Social media bio (yes, these bio pages get indexed and cited by LLM chatbots)
After just optimizing these pages, we have seen a significant change in how LLMs represent our brand.
Here’s the screenshot of Perplexity before optimizing our website:
Here’s the screenshot of Perplexity after 1 week of optimizing our website and pages that we have control:
While writing content for these pages and social media bio, be careful about the sentence structure.
Search engines use NLP to crawl, index, and rank content by analyzing its relevance, structure, and semantics.
Your content should use short, simple, and direct sentences, which are easier for NLP models to process and understand. For instance:
“Rentpost is a property management software…”
"Tenants can pay rent via credit card or bank account."
This clear structure avoids ambiguity, making it straightforward for tokenization, dependency parsing, and semantic analysis.
Examples:
❌ Not recommended
✔️ Recommended
If you’re still not sure how sentence structure can affect how machines can understand your page?
Here’s a comparison of the above two content examples (analyzed topic relevance using Text Razor):
The well-optimized content is more focused on property management, whereas the generic one has mixed intent and relevance.
Action items:
Revisit each social media bio and optimize it with the right keywords. Ensure the structure is simple, direct answer type, and optimized for NLP.
Likewise, optimize your landing page and homepage content (including heading structure)
IMPORTANT TIP:
Create a dedicated FAQ page to answer all brand-related queries.
The goal here is to feed search engines and LLMs enough information about your brand and its positioning.
Check Google Search Console, and create a list of all branded questions (don’t forget to add the question ‘What is Product X?’ 😉)
Step 4. Reverse engineer your competitors
Remember in the first step, we created the list of all sources used by LLM chatbots?
Now, it’s time to get featured on all of those pages.
Is it a listicle page? Outreach or sponsor to get featured in the list
YouTube videos used as sources in LLMs? Ask other YouTubers to do product reviews on their channels
Does your competitor have a Wikipedia page? Try to get one too.
Your competitors are ranking on top of directories? Get more reviews for your brand and try to get on the top of those directories too.
In our case,
acquired more user reviews on G2, Capterra, etc.
started creating YouTube videos on our channel and other channels
Trying to get a Wikipedia page (outsourcing to an agency)
Outreaching to all listicle articles to get our brand featured on the list
Maintaining social media presence through consistent content distribution
Step 5. Get mentioned in pages where your competitors are already listed
Your goal should be to feed enough information about your brand from various sources.
This is why working on your website alone is not enough to succeed in SEO (and LLM models).
One quick hack is to use Reddit and Quora:
Google: [Your niche topic] + site: reddit.com/ [Your niche topic] + site: quora.com
There are dozens of questions on these UGC platforms where you can simply mention your brand name.
I already have a video on finding questions using Scraper:
NOTE: Start finding the Reddit/Quora questions from Google search (to ensure these pages are already indexed) — this will help in AI overviews as well.
Step 6. Use UGC platforms for content distribution
During our research, we found that LLMs often cite LinkedIn and Medium articles as sources.
So, why not use the same platforms to distribute your listicle articles?
Here’s what we planned:
Consistently publishing listicle articles on these UGC platforms (there are more than just LinkedIn and Medium)
Mention your brand in the list
Don’t avoid mentioning your competitors in the list as well
Step 7. Publish linkable assets
As an agency, we always publish linkable and research-driven statistics pages biweekly.
Previously we used to do this for passive link-building purposes.
However, we now found that these data statistics can be cited in LLM chatbots often— improving your share of voice in these platforms.
One thing we are yet to start with:
PR campaigns to publish brand story and its positioning (will share more on this once we see positive results with PR campaigns)
Step 8. Improve your organic ranking in Google
It’s difficult to optimize for LLM when your website doesn’t rank for any business-relevant keywords.
A study from Seerinteractive shows that organic rankings can greatly influence your brand presence in LLM chatbots.
So, do your basics of SEO— quality content, better content experience, technical health (smooth crawling and indexing), backlinks.
Let's make SEO the biggest growth engine for your brand.
We started Accrue SERP with one clear goal— making SEO an outcome-based marketing channel (leads, signups, & conversions) rather than chasing site traffic that doesn’t add any business value to your bottom line.
Let’s schedule a 15 to 30-minute meeting to understand how we can help you achieve business growth with our outcome-based SEO services in India.
Let’s connect on LinkedIn.
Thanks a million for this reading! It is gold from beginning to end!