Optimize for LLMs & AI with Third Party Signals

We earn commissions if you shop through the links below.

Optimize for LLMs & AI with Third Party Signals

This post is about how multiple marketing and branding teams can work together to get featured in LLM results like OpenAI’s ChatGPT and Perplexity by using third party signals.  It includes being a recommended brand in the lists of where to shop and recommended stores based on what we’re seeing.  It is not about what to do on your own website or to use a short cut like clicking thumbs up on outputs being good (which is bad advice and trackable), there’s enough of those posts already.

TL:DR

When optimizing third party signals for LLMs, backlinks likely have little to no impact, but the content your affiliate, PR, influencer, media, and other teams get out in the world likely does if the LLM or AI is sourcing information from the content you partner with.  If you get these teams working together vs. working in silos, AI optimization may become a lot easier and you’ll be setting yourself up for success in the long run.

For the short version, read the paragraph that starts with “Leaders that…” and then skip to the paragraph that starts with “When you go into an LLM.”  You will miss a bunch of things that I think are important, but also get the gist of the post.

The Full Post

I’m seeing claims that link building affects AI results including LLMs like Google AI Overviews and ChatGPT.  While there is merit in the thought that if links can be “crawled” (spiders don’t actually crawl through and follow links in a literal sense) this could be a thing.  Right now it likely is not.  However, having an affiliate program with affiliate links may have an unexpected beneficial impact on LLM and AI optimizations.

This can be especially true if the lists are being used for information and trust by the LLM.  The amount of lists you’re on with the content surrounding the inclusion may also have an impact.  As some of our clients get featured in themed shopping lists and “best places to” lists, we’re seeing them show up more often in recommended brands, products, and stores by the LLMs AI results.  We also have clients that terminated multiple relationships and they disappeared as the websites being used as sources removed them from their content or reduced the frequency of inclusions, positions on lists, and sentiment.

I’ve seen zero documentation that backlinks are part of what AI is using as a trust signal like SEO algorithms do.  I do see how link building from being in an article on a site that LLMs trust as a source of information can send a trust and credibility signal to the LLM.  These lists can also send information about the brand to the AI part of the LLM.

It isn’t the link from the article helping with AI or LLM results like in SEO, it is the existence of the website or brand within the content that the LLM trusts and sources their information from that gives the AI overview or LLM feature.

If this is true, the more trusted and likely topically relevant resources your brand is featured in, the more likely you are to be featured in an LLM like ChatGPT or Perplexity result.  There are other factors and trends I’ve noticed, but nothing I feel confident sharing yet.  It does not mean I’m right, just what I’m observing.

I do believe that having an affiliate program makes it easier to show up in AI and LLM rankings like ChatGPT and Perplexity because you get featured in “trusted resources” more easily and more often.  Media companies for example feature brands with affiliate links in shopping lists more often than including brands without affiliate programs as there’s less money to be made.

I also think the LLMs will catch onto these “trust signals” and devalue affiliate content and listcicles built for revenue vs. helping the user.   Similar to how search engines have begun devaluing lists built to make money vs. help a user make an educated purchase.

Leaders from AI and LLM companies like ChatGPT’s Sam Altman have said they will likely take affiliate commissions from links within results. I’ve already shared how search engines like Yahoo and Bing are using affiliate links on SERP page features like some knowledge panels, so why would this be different?  Every company has a right to make money and monetize their own traffic, LLMs included.

Microsoft has a merchant program with Copilot and Perplexity has multiple shopping experiences including shop like a pro which is like an affiliate toolbar times ten.  It has the power to provide informative solutions and drive new customers and top funnel sales to brands.  It is the perfect opportunity to monetize through affiliate links.  I’m absolutely loving it as a consumer.

As a creator I don’t love it because it can take my content and use it to make money by using my knowledge and not allowing users to shop through my affiliate links or send me traffic for CPM ad revenue.

One thing that would be interesting to test and measure is if companies who just launch an affiliate program, who also have trust with the AI algorithm already, start getting more features in LLM results as the LLM is making money from it via the affiliate program.  But that isn’t the topic of the post.  This post is about optimizing for AI and LLM results via off site metrics and marketing channels.

When you go into an LLM or AI result, you can find the sites and sources by asking for citations.  I include some examples below.  It isn’t any one channel in particular that can impact your brand being featured, it is the marketing channels and branding working together towards the same goal.  Multiple teams work with content creators that can be indexed, cited, or crawled for information including:

  • PR
  • Affiliate
  • SEO (link builders)
  • Branding
  • Media buyers
  • Influencer and ambassador managers

Each one can help the company to show up better in LLMs, but when the teams work as a silo vs. together, everyone fights to take credit and no body wins.  In theory SEO should win because it is a similar optimization, but it could have been your affiliate and media buying team and SEO had zero influence.  Here’s a few examples of how multiple teams can work together so that your brand may show up more often in LLM or AI search results via third party signals.

SEO and Affiliate

SEO and link builders love pitching media companies, but media companies including blogs and mass media outlets have begun monetizing through affiliate links.  This means the backlinks are no longer there because of affiliate marketing, but the SEO or link builder pitched and got the placement.

Without the affiliate program, the link placement would likely not have been as successful.  While the link builder is taking credit, so is the affiliate manager as sales and clicks show up in their channel.  This throws off attribution.  Although there is little to no SEO benefit here as affiliate links are not backlinks for SEO, being featured on these lists helps brands get featured in LLMs when someone asks for “the best XYZ product” or “which XYZ to buy” if these sites are being cited and sources for information.

SEO and affiliate have now played a major part in securing the LLM and AI optimization and both channels should get credit for the media placement and the LLM result.

In this first screen shot I asked ChatGPT 40 for “the best places to buy blue jeans” and to share where they found the information.  An overwhelming majority of the sources are affiliate listcicles and niche websites.  I’ve seen first hand that once our clients brands are featured in affiliate listcicles and niche sites that get crawled, the LLMs start including them in the results.

This result includes People magazine, RealSimple, NY Magazine multiple times, a niche site, etc… which you can see below.  I’ve also had clients terminate multiple affiliate relationships and observed them being featured less often and now excluded from some of these lists.

affiliate listcicles for LLM and AI optimization

One thing to keep in mind is shopping queries are different.  One word can change the intent of the search from informative to conversion and also the outputs of the results.  Reddit may come into play with brand’s that are trusted, have shopping feeds, etc… Instead of a list of brands, it shows brands and products directly.

If you add “and the reasons why” and “please share the sources” or “please citate your sources”, you’ll see where and how it is sourcing information for the result.  Now you can build relationships with these site or be more active in the communities and forums.

In this scenario affiliate may not play a part of the brand showing up unless the affiliate links become part of the revenue model and ChatGPT no longer shares stores that they cannot monetize.  If that isn’t the case, affiliate can still play a part because when the trusted content sites that use affiliate links all feature the brand, the LLM may feel the brand is trustworthy and feature it in the shopping grid mix.

I clicked multiple URLs below and there were zero affiliate links from the ChatGPT results, but multiple do have ChatGPT parameters for tracking consumers.

AI shopping results no affiliate links

PR, Affiliate, and Paid Media

Sometimes LLM optimizations won’t be SEO at all.  Instead PR pitches the outlet, but the affiliate program opened the door for them to get the inclusion.  In addition, the media buyers paid for coverage via an advertorial or guaranteed placement in a listicle.  This combination of events causes the brand to get featured in an LLM output, but who gets credit?

The media team did a media buy and paid for guaranteed space.  This space was strategic in the listicle, and this listcicle is sourced by multiple relevant LLM and AI answers.  Affiliate managers will want to take credit because the affiliate link is what is used in the article that lead to the LLM output regardless of the media buy or PR parameter.  SEO will also take credit because it is in the LLM result and SEOs job is becoming part search engine and part LLM.  Media will want credit because their budget paid for the feature, and PR will say it was a pickup because they are responsible for the features.

All channels played a role here.  Affiliate built the relationship with the media company for features outside of media buys.  Because of the affiliate program, PR can get more placements more easily.  Media buyers can guarantee more and better spaces, sometimes at reduced costs because of commissions, and this is because of the affiliate team.

All of these lead to the trust signals that helped the LLM trust the brand and listicles as a place to shop from.  No one team should get credit, but each can work on a strategy together.

  1. SEO shares which sites are being cited and sourced more often for shopping and conversion queries.
  2. Affiliate builds relationships with the media team and gets pay to play placement information.
  3. PR tries a pitch to see if an addition can be made to specific current lists without having to pay on top of affiliate commissions.
  4. If PR cannot make it happen, affiliate should try next.  Chances are it will become a pay to play to get featured and that is when media comes in to buy the space.
  5. Affiliate gets on a call with Media to get the budget and the two negotiate with the outlet for a hybrid media buy.
    1. Fee + commissions
    2. CPC on the affiliate link + commission + media fee
    3. CPC + commission
  6. Now the company has the placement, the LLM finds it and adds a trust signal to the brand, and may show the brand in new outputs or results.

One other thing to consider are the deals AI companies are cutting deals with ecommerce providers.  A rumor that has credibility (I can’t find it on the OpenAI site yet) is that ChatGPT and Shopify are working on a shopping experience inside ChatGPT.  If the big shopping cart companies close out the small ones, and LLMs take off, it could disadvantage custom shopping carts as there is no relationship.  At the same time there could be advantages to smaller brands that invest in the shopping cart systems used by the LLMs.

Influencer and Ambassador

Influencer and ambassador can also impact a brand showing up in LLM results.  Both work with content creators, and LLMs like ChatGPT may use their content to determine if the brand is trustworthy, and what the current sentiment or public opinion is of that brand or specific product.

Instead of repeating all of the things from above, here is how ChatGPT says it uses social media influencers in making decisions.  If this is true, launching a large influencer campaign may be able to send a trust signal for your brand or product and get it featured in LLM search results.

LLMs and AI using social media for rankings and listings

Most marketing channels are a mix of multiple teams which is why attribution is never easy. When optimizing third party signals for LLMs, backlinks likely have little to no impact, but the content your affiliate, PR, influencer, media, and other teams get out in the world likely does.  If you get these teams working together vs. working in silos, AI optimization may become a lot easier and you’ll be setting yourself up for success in the long run.

Join My Newsletter & Never Miss Another Post!

Contact Us

Contact Us

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Scroll to Top