SEO is the design language for problems we solve

I was on Edward Sturm’s SEO podcast last week and it made me reflect on how much the role of SEO has shifted over the last couple of years. For a long time SEO was treated almost like a technical discipline inside marketing. You optimized pages, built links, published content, and gradually rankings improved.

But discovery works differently today.

Search engines are still important, but large language models are becoming another layer on top. When someone asks a question in an LLM, the system isn’t simply scanning for technically optimized pages. It’s looking for sources that consistently demonstrate authority on the topic.

That leads to a very different question for marketers.

Not “how do we rank for this keyword?”

But: does the market see us as one of the companies that actually understands this problem?

Topical authority

A lot of the podcast discussion centered on how this played out at Orb.

Before Orb I worked at Instabase, where the product space was extremely broad. AI workflow automation touches many industries and workflows, which meant the SEO strategy could also be broad. The objective was simply to create many entry points into the site and capture traffic across a wide range of problems.

Orb is very different. The category is more narrow: usage-based billing and monetization infrastructure for SaaS and AI companies.

That forced a different approach. Instead of going wide, we focused on going deep around a small cluster of closely related topics: billing, pricing, monetization models, packaging, invoicing. The objective wasn’t simply to rank for “usage-based billing.” It was to make sure that anyone researching modern monetization models would repeatedly encounter Orb’s thinking.

The signal that unlocked much of this didn’t come from keyword tools.

It came from conversations.

At conferences we would have “usage-based billing” on our booth banners. People would walk up, see the message, and almost immediately start talking about pricing. How should we price AI products? How do we package usage? How do we avoid losing money if compute costs spike?

That was a clear signal. The conversation the market wanted to have often started with pricing, not billing.

So we leaned into that. One of the formats that ended up working particularly well was analyzing how large SaaS companies structure their pricing pages. Not product marketing — simply breaking down how different companies package and monetize their products.

Those articles became entry points for real buying journeys. Someone discovers Orb through a pricing analysis article. Two months later they return through an ad. Then they explore product pages or case studies. Eventually they book a meeting.

If you looked only at the first interaction, it looked like informational traffic. But when you mapped the full journey, the pattern became obvious. Some of our highest-value opportunities started with what looked like purely educational content.

You simply cannot be data driven without data. Sometimes the fastest way to learn is to publish something, see how the market reacts, and iterate.

New role for brand

Another theme that came up on the podcast is how closely brand and SEO are now connected.

For years brand marketing and performance marketing were treated as separate activities. One was measurable, the other less so. In practice they reinforce each other constantly.

If people begin to associate your brand with a specific concept, search behavior reflects that. At one point we explicitly had “usage-based billing” in the H1 of our homepage to make the connection obvious. Eventually we removed it because it wasn’t necessary anymore.

The association had already formed.

Orb started ranking independently for both “Orb” and “usage-based billing” because enough signals across the market connected the two. That wasn’t just technical SEO. It was the narrative showing up in search behavior.

SEO works best when it becomes the technical reflection of your broader marketing narrative.

Have fun with it

The most fun example we talked about on the podcast was something slightly unconventional we built on our pricing page.

There’s an Easter egg called X-ray Vision. If you click it, the page reveals an annotated version of our pricing explaining exactly why it’s structured the way it is.

It’s essentially a transparent teardown of our own pricing model.

We didn’t build it as a conversion tactic. We built it because many visitors were clearly trying to understand how to price their own products. Instead of hiding our thinking, we decided to show it.

Pricing experts loved it. People shared it. It generated discussion.

And it illustrates something important about discovery today. Content that demonstrates real thinking about a problem tends to travel further than content that simply promotes a product. It’s also exactly the kind of material algorithms and LLM systems tend to surface.

Be relevant

If there’s a broader takeaway from the conversation, it’s that search is becoming less mechanical.

Search engines and language models are both trying to answer the same underlying question: who actually understands this topic?

If your company consistently contributes useful thinking around a problem space, that authority accumulates. It shows up in search rankings, in AI responses, in branded searches, and in how people talk about your category.

Which is why I increasingly think of SEO less as a standalone channel and more as a signal.

A signal of whether your company is actually shaping the conversation around the problem it solves.

The goal is not just to rank for a keyword. The goal is to become part of how the market understands a problem.

If that’s happening, the rankings tend to follow.