What did Profound raise, and why should a small team care?
Profound raised a $180 million Series D at a $1.8 billion valuation, announced on 15 September 2026 and co-led by Sequoia Capital and Kleiner Perkins (Profound's announcement). It came less than seven months after Profound's $96 million Series C. Existing investors Lightspeed Venture Partners, Khosla Ventures, Saga Ventures, Evantic and South Park Commons also took part.
Profound builds marketing software that shows brands how they appear in AI answers. TechCrunch placed it in a wave of GEO and AEO startups (generative engine optimisation and answer engine optimisation): companies trying to get products surfaced in the AI systems customers increasingly use for search. Profound says it serves over 1,000 enterprise brands, including Comcast, Walmart and The Estée Lauder Companies, and TechCrunch reports the company says its revenue increased 3x in the past six months.
That is enterprise software. As of 20 October 2026, Profound's pricing page lists a free trial for brands, with 50 prompts run daily for 7 days, and a custom-priced Enterprise plan, with no public price. The useful part for a three-person team is the list of things these tools measure, because most of it can be checked by hand once a month.
What do AEO tools measure?
According to the vendors' own pages, AEO tools measure how often AI assistants name your brand, which websites those answers cite, and how you compare with competitors. Profound's Answer Engine Insights page lists visibility scores and share of voice, brand sentiment and themes, citation sources and their authority, and competitor rankings. Here is what each one means and how to approximate it for free:
| Metric | What it tells you | Free way to check |
|---|---|---|
| Visibility score | How often your brand appears in AI answers for the questions you track | Questions where you are named, out of 20 |
| Share of voice | Your mentions compared with competitors' mentions | Count every brand named across all answers |
| Citation sources | Which websites the answers draw on | Log every site an answer links to or cites |
| Sentiment and themes | How the assistant describes you | Note the words used about you, and anything wrong |
| Competitor ranking | Who is named first and most often | Record the order brands appear in |
| Prompt volumes | What people actually ask AI assistants | No way to collect it by hand; use your own sales and support questions |
| AI crawler and referral traffic | Which AI bots read your pages, and how many people click through | Server logs and your analytics referral report |
Two rows need data that asking the assistants can't give you. Profound says its Prompt Volumes data comes from conversations it licenses from double-opt-in consumer panels, scaled with statistical modelling. Its Agent Analytics product uses server logs to see which AI crawlers visit a site. The rest is a structured way of asking questions and writing down the answers.
One caution applies to any manual check. Profound's page says: "AI platforms don't return the same answer twice." It runs every tracked prompt daily for that reason. A monthly check by hand is a sample, so treat one month's change as a hint, and three months moving the same way as a trend.
What do AEO tools cost for a small team?
Of the tools below, the entry-level paid plans cost $25 to $99 a month for 15 to 25 tracked prompts, while Profound lists a free 7-day trial and a custom-priced Enterprise plan. Prices come from each vendor's own page, checked 20 October 2026.
| Option | Price | Prompts tracked | Assistants covered |
|---|---|---|---|
| Profound Trial | Free | 50 prompts, run daily for 7 days | ChatGPT, Gemini, Google AI Overviews |
| Profound Enterprise | Custom | Tailored prompt tracking plan | Up to 9 answer engines |
| Otterly.AI Lite | $29 a month billed monthly, or $25 a month billed annually | 15 prompts, tracked daily | ChatGPT, Google AI Overviews, Perplexity, Microsoft Copilot; Gemini is an add-on |
| Semrush AI Visibility Toolkit | $99 a month, no free trial | 25 prompts in Prompt Tracking | Semrush names Google AI Mode and ChatGPT among the platforms |
| The manual routine below | $0 | 20 questions, once a month | ChatGPT, Gemini, Perplexity |
The manual routine produces 60 answers a month (20 questions × 3 assistants). Otterly's Lite plan tracks fewer questions but runs them every day, so you see a daily view instead of a monthly sample. Start by hand. Pay for a tool once the monthly check takes longer than it is worth, or once you need daily data.
How to run a free monthly AEO check with three people
This is our suggested routine, not a vendor method: run it as one block of time each month, with each person owning one assistant and asking it the same 20 buyer questions on the same day. If you have no marketer, this fits the approach in our guide to marketing without a marketing team: a small, fixed routine done every month.
Write the 20 questions your buyers ask
Pull them from sales calls, demo notes, support emails and questions prospects ask before they buy. Use four groups of five: the problem (how do I...), the category (best tool for...), comparisons (your competitor vs another, alternatives to a competitor) and your brand (what is your company, is it good for...). Write them the way a buyer would type them, and keep the wording identical every month.
Give each person one assistant
One person runs ChatGPT, one Gemini and one Perplexity. Ask each question in a new chat, signed out or with memory turned off where the assistant allows it, so each run starts from the same settings. Run all 20 on the same day.
Log every answer
For each answer, record which brands are named and in what order, which sites are linked or cited, whether any of your own pages appear, and anything it gets wrong about you. Perplexity's help centre says each answer includes numbered citations. Google's Gemini help page says not all Gemini responses include sources, so log those answers as having no sources.
Score the month
Work out three numbers per assistant. Mention rate: questions where you are named, divided by 20. Citation rate: questions where one of your pages is cited, divided by 20. Share of voice: your mentions divided by all brand mentions. Put them next to last month's.
Pick three fixes
Choose the three questions where a competitor is named and you are not, and act on them before next month's check, using the fixes further down.
AEO log template: copy this table
Use one row per question per assistant, in a shared spreadsheet with one tab per month. The first row shows the format only; every value in brackets is a placeholder for what you actually see.
| Date | Assistant | Question | Brands named, in order | Are we named? | Sites cited | Our pages cited | Wrong about us | Action |
|---|---|---|---|---|---|---|---|---|
| [date] | [assistant] | [question, exact wording] | [brand 1], [brand 2], [brand 3] | [yes, position / no] | [domains] | [URLs / none] | [claim / none] | [next step] |
At the bottom of each month's tab, add a summary:
| Metric | Formula | ChatGPT | Gemini | Perplexity |
|---|---|---|---|---|
| Mention rate | Questions where we are named ÷ 20 | |||
| Citation rate | Questions citing one of our pages ÷ 20 | |||
| Share of voice | Our mentions ÷ all brand mentions | |||
| Most cited site | Domain that appears most in Sites cited | |||
| Errors about us | Rows with something wrong |
After three months the log answers two practical questions: which assistant matters most for your category, and which third-party sites keep getting cited when you don't.
What should you fix when AI assistants leave you out?
Fix access first, then the pages that answer buyers' questions, then the third-party sites the assistants already cite.
- Check your robots.txt for AI search crawlers. OpenAI's crawler documentation says sites that opt out of OAI-SearchBot will not be shown in ChatGPT search answers, though they can still appear as navigational links, and that robots.txt changes take about 24 hours to apply. Perplexity's crawler page recommends allowing PerplexityBot to appear in its search results; changes can take up to 24 hours.
- Skip special files for Google. Google's guide to AI features (last updated 10 December 2025) says there are no additional requirements to appear in AI Overviews or AI Mode, and no need for new machine-readable files, AI text files or special schema.org markup. A page must be indexed and eligible to show in Search with a snippet. Search Console includes traffic from these features in its overall figures, in the Performance report under the Web search type.
- Answer the exact question on your own site. If buyers ask for alternatives to a competitor and only the competitor's pages exist, write a fair, dated comparison. Siela's SEO Blog Writer drafts posts like this from an outline you approve, checking the top 10 Google results and People Also Ask first; it does not publish.
- Work on the sites that keep getting cited. When the same review site, directory or community thread appears across answers, make sure your listing there is complete and current.
- Trace every error to its source. When an answer gets something wrong about you, open the pages it cites and find where the claim came from. Profound sells this as a feature (FactCheck); by hand, it means correcting your own page or asking the site owner to update theirs.
When is a paid AEO tool worth it?
A paid tool is worth it when you need daily data, more assistants or prompt volumes, none of which a monthly check by hand provides.
- More questions or assistants. Past 20 questions across three assistants, the manual log gets slow. Otterly's Lite plan covers 15 prompts across four AI search engines daily; Profound's Enterprise plan tracks up to 9 answer engines.
- Daily movement. Around a launch or a pricing change, a monthly sample is too coarse.
- Real demand. Knowing which questions people ask most needs panel data like Profound's Prompt Volumes, which you can't gather yourself.
Siela's Competitor Watch is narrower: a weekly briefing every Monday morning on what each competitor launched, priced, hired or posted, which also checks AI-assistant visibility. Whichever route you choose, keep the 20-question log. It costs nothing but time, and it records what buyers are told month by month.
Frequently asked questions
What is AEO for a small business?
AEO, or answer engine optimisation, is the work of getting your business named and cited when people ask AI assistants such as ChatGPT, Gemini and Perplexity for advice. For a small business it can start with a free monthly check: ask the 20 questions your buyers ask, record which brands and sites appear, then fix the gaps on your own site and on the sites the assistants cite.
How much did Profound raise in its Series D?
Profound announced a $180 million Series D at a $1.8 billion valuation on 15 September 2026, co-led by Sequoia Capital and Kleiner Perkins. It came less than seven months after a $96 million Series C. Profound says it serves over 1,000 enterprise brands, and on 20 October 2026 its pricing page for brands listed a free trial and a custom-priced Enterprise plan.
Do I need special files or schema to appear in Google AI Overviews?
No, according to Google. Its guide to AI features, last updated 10 December 2025, says there are no additional requirements to appear in AI Overviews or AI Mode, and you don't need new machine-readable files, AI text files or special schema.org markup. A page must be indexed and eligible to appear in Google Search with a snippet.
What is share of voice in an AEO check?
In a manual AEO log, share of voice is your brand's mentions divided by all brand mentions across the answers you track. Profound's Answer Engine Insights page lists share of voice alongside visibility scores, citation sources and competitor rankings. By hand, count every brand named across your 20 questions on each assistant, divide your count by the total, and compare it with last month's figure.
How often should a small team check its AI visibility?
We suggest once a month for a check done by hand, using the same questions each time. Profound says AI answers vary from run to run, which is why it runs every tracked prompt daily. Our advice: treat a single month's change as a hint, and wait until about three months move in the same direction before changing your plan.
Sources
- Profound: Profound Raises $180M Series D at $1.8B Valuation (15 Sep 2026; checked 20 Oct 2026)
- TechCrunch: AEO startup Profound hits unicorn valuation, raises $180M Series D (15 Sep 2026; checked 20 Oct 2026)
- Profound pricing (checked 20 Oct 2026)
- Profound: Answer Engine Insights (checked 20 Oct 2026)
- Profound: Prompt Volumes (checked 20 Oct 2026)
- Profound: Agent Analytics (checked 20 Oct 2026)
- Otterly.AI pricing (checked 20 Oct 2026)
- Semrush Knowledge Base: AI Visibility Toolkit (checked 20 Oct 2026)
- OpenAI: Overview of OpenAI crawlers (checked 20 Oct 2026)
- Perplexity: Perplexity crawlers (checked 20 Oct 2026)
- Perplexity Help Center: How does Perplexity work? (last modified 28 Aug 2026; checked 20 Oct 2026)
- Gemini Apps Help: View related sources from Gemini Apps (checked 20 Oct 2026)
- Google Search Central: AI features and your website (last updated 10 Dec 2025; checked 20 Oct 2026)