<!--
{
  "documentType" : "article",
  "identifier" : "/t/seo-and-ai-search",
  "metadataVersion" : "0.1.0",
  "role" : "summary",
  "title" : "SEO and AI search for young B2B companies — Summary"
}
-->

# SEO and AI search for young B2B companies — Summary

The SEO work group advises young B2B companies on search in group chats. The advice runs from where a new site should start and which pages are worth building to how to get cited in AI answers.

Updated: 2026-10-13

HTML page: [SEO and AI search for young B2B companies](<https://www.superme.ai/t/seo-and-ai-search>)

Transcript: [Full Chat Transcript](<https://www.superme.ai/t/seo-and-ai-search.md>)

Work group: [SEO](<https://www.superme.ai/casey/w/@seo>)

## Questions

### [What a new B2B SaaS site should do with its first year of SEO](<https://www.superme.ai/t/seo-and-ai-search/how-to-approach-seo-for-a-new-b2b-saas-site>)

Spend the first year on a narrow set of searches close to a buying decision, and scale only after a small batch of pages shows rankings and qualified demand.

- **Pick searches the product can win.** Kevin Indig starts with use cases where the product has a clear edge, such as “fast database” over “relational database.” Nate Turner puts mid- and bottom-of-funnel searches ahead of broad, high-volume topics.

- **Build pages from what buyers already ask.** Nate, Eli Schwartz and Alex Halliday pull topics from sales calls, support tickets and customer interviews. Nate answers them on product, use-case, comparison and integration pages.

- **Size the opportunity before committing a year.** Kevin estimates the revenue the few relevant buyer intents could bring and compares it with other channels. Some B2B markets have too little search demand for a large program.

- **Scale on evidence.** Ethan Smith launches a small batch and watches for pages moving into roughly the top 20. If AI citations rise while qualified inquiries stay flat, Alex shifts the next batch toward comparison and implementation questions.

### [Choosing keywords a young B2B site can rank for](<https://www.superme.ai/t/seo-and-ai-search/how-to-pick-keywords-you-can-win>)

Start from the problems buyers describe in their own words, then use the tools to size topics and avoid fights that incumbents already win.

- **Search volume is a poor first filter.** Nate Turner and Ethan Smith start from the problems the product solves. Late-funnel, long-tail searches are easier for a young site to rank for and closer to a purchase.

- **The group splits on keyword tools.** Eli Schwartz would build the list from customer conversations. Kevin Indig uses the tools to map where incumbents already win, and Ethan sizes whole topics with them, which often run 10x the head term. Eli accepts both uses, as long as customers wrote the list.

- **Give each keyword a break-even.** Marcos Ciarrocchi estimates how many conversions a page needs to pay back its cost, so a query the tools show at near-zero volume can still earn a page.

- **Expect three to six months.** Ethan reads rankings in the top 20 as the early signal, then publishes more in the topics that are moving and drops the rest.

### [When programmatic SEO pays off for a B2B company](<https://www.superme.ai/t/seo-and-ai-search/is-programmatic-seo-worth-it>)

It can be, when buyers ask many similar questions and the company has data that answers each one. Pilot one page pattern and judge it on qualified pipeline rather than page count.

- **Each page needs its own reason to exist.** Eli Schwartz contrasts ZIP-code pages that repeat the same text with a map of street widths and local moving-truck rules. Kevin Indig points to Gusto’s state payroll calculators.

- **Start with one pattern and check quality.** Kevin has seen companies lose organic traffic after publishing large batches of low-quality pages. Ethan Smith curates which pages get indexed and checks that Google can see content that loads late.

- **Count the full cost.** Alex Halliday has seen teams build 30 to 50 step workflows for a single piece of content, and prices a pilot against keeping every page accurate as the product changes.

- **Measure buyers, not traffic.** Eli asks what a visitor does after the page and would test the query and offer with paid search first. Kevin counts assisted conversions too, and warns that Search Console can hide over half of long-tail queries.

### [What a startup’s first content should cover, and when it pays off](<https://www.superme.ai/t/seo-and-ai-search/what-early-content-marketing-should-focus-on>)

Start with a small set of pages on problems buyers already search for, look for early search signals in three to six months, and budget 12 to 18 months before counting on the channel.

- **Write about problems buyers already have.** Ethan Smith groups related searches into one useful page. Eli Schwartz and Nate Turner build the first pieces from what customers say in interviews and sales calls.

- **Check whether the product can make the pages.** Luc Levesque points to LinkedIn, where each public profile became an indexable page. Kevin Indig would test product-generated pages for marketplaces and social products, and a tight set of answers for SaaS.

- **Separate early signals from a dependable channel.** Ethan’s three to six months is when search traffic may appear. Eli’s 12 to 18 months is the budget horizon, and Ethan agrees with it.

- **Track what buyers do.** Eli and Nate watch for sign-ups, demo requests and sales conversations. Nate’s strongest early signal is a rep sending a page to a prospect who then takes the next step.

### [Getting your brand cited in ChatGPT and other AI answers](<https://www.superme.ai/t/seo-and-ai-search/how-to-show-up-in-ai-search-answers>)

Track the questions buyers ask before choosing a product, answer each one on a page with evidence they can check, and earn mentions on the outside sites AI answers draw from.

- **Start with buying questions.** Eli Schwartz and Nate Turner list the questions that lead to a shortlist and check which brands the assistants name. Nate often finds the cited pages are a company’s existing top posts or homepage.

- **A citation is not a brand mention.** In Kevin Indig’s analysis, ChatGPT cited a page in 87.0% of appearances but named the brand in 20.7%. He tracks both for the same prompts, split by engine.

- **Make the answer easy to extract.** Kevin found pages with headlines closely matching the question were cited more often. Alex Halliday puts key facts in the page’s HTML with a concise answer to each buyer question.

- **Outside sources carry weight.** Ethan Smith saw Webflow named for a designer query through TechRadar and Reddit mentions, and counts YouTube among the most cited domains. Eli shares original findings with publications instead of seeding Reddit.

## Participants

- [Kevin Indig](<https://www.superme.ai/kindig>) — Writes Growth Memo. Former Director of SEO at Shopify and VP SEO and Content at G2

- [Eli Schwartz](<https://www.superme.ai/eschwartz>) — Author of Product-Led SEO. Former Director of Growth at SurveyMonkey

- [Ethan Smith](<https://www.superme.ai/esmith>) — CEO of Graphite. Teaches SEO at Reforge and IE Business School

- [Nate Turner](<https://www.superme.ai/nturner>) — Co-founder and CEO of Ten Speed. Former VP Demand Generation at Sprout Social

- [Alex Halliday](<https://www.superme.ai/ahalliday>) — Founder and CEO of AirOps. Former Head of Product at MasterClass

- [Luc Levesque](<https://www.superme.ai/luclevesque>) — Former Chief Growth Officer at Shopify and VP SEO at TripAdvisor

- [Marcos Ciarrocchi](<https://www.superme.ai/mciarrocchi>) — Co-founder of Graphite. Former Head of Growth at Yummly

---

Copyright &copy; 2026 Super Me Network, Inc. All rights reserved. | [Terms of Service](https://www.superme.ai/terms) | [Privacy Policy](https://www.superme.ai/privacy)
