The GEO/AEO Market in 2027: A Forecast
ChatGPT use in the U.S. has more than doubled since 2023, and two research firms now put the GEO/AEO services market on very different 2031 paths.
Updated August 27, 2026
Three years ago GEO was a conference paper. This year Pew can report that 44 percent of U.S. adults have used ChatGPT, and firms that optimize for AI answers are showing up as a line item in market reports. The two houses that published a TAM do not agree. The survey data is harder to argue with. We have been watching this from a directory that reviews the shops doing the work. The numbers below are theirs. The forecast at the end is ours.
Americans Started Searching In Chatbots
In February 2026 Pew Research Center asked 5,119 U.S. adults whether they had used ChatGPT. The share who said yes had climbed from 18 percent in 2023 to 23 percent in 2024, 34 percent in 2025, and 44 percent this year. The fieldwork ran Feb. 17-23.
Chatbots as a class moved with it. About half of U.S. adults, 49 percent, now say they have used ChatGPT, Gemini, or Copilot, up from 33 percent in 2024. Forty-two percent use them to search for information. Sixty percent read the AI summary at the top of a search page. A quarter, 24 percent, open a chatbot every day.
Our read is that chatbot search is already a mass habit in the U.S., not a hobby among marketers. Budgets that still treat GEO as experimental are late.
Stanford's AI Index put the same curve in a longer frame. Generative AI reached about 53 percent population-level adoption within three years of mass-market introduction, faster than the personal computer or the internet at the same stage. The Index also cites Bick, Blandin, and Deming for an estimated $172 billion a year in U.S. consumer surplus from generative AI tools by early 2026. People are not waiting for a category name to settle.
Pew Research Center, 2026. The 2026 question also asked about other chatbots.
| Year | Share who have used ChatGPT (%) |
|---|---|
| 2023 | 18 |
| 2024 | 23 |
| 2025 | 34 |
| 2026 | 44 |
Pew Research Center, 2026.
| Task | Share of U.S. adults (%) |
|---|---|
| Search for information | 42 |
| Work tasks among employed | 38 |
| Fun | 25 |
| Images/video | 24 |
| Medical advice | 20 |
The Click That Does Not Happen
Aggarwal and colleagues named the practice at KDD in 2024. On a benchmark they called GEO-bench, black-box edits to a page raised visibility in generative-engine answers by as much as 40 percent. That is the upside vendors sell. The rest of the record is about how thin the prize is once the answer is already on the page.
Who actually gets cited is a narrower fight. At SIGIR 2026, Vishwakarma, Kumar, and Jamidar ran 252,000 paired trials across six language models. Topical relevance and list position decided who was cited first. Price facts and recent timestamps helped, consistently. The models still reward being the obvious source, not the loudest one.
Google's AI Overviews make the click itself scarce. Chapekis, Lieb, Shah, and Smith watched a representative panel of about 900 U.S. adults. People clicked a search result on 15 percent of visits when there was no Overview and on 8 percent when there was one. They clicked a citation inside the Overview on about 1 percent of those visits. They ended the session on 26 percent of Overview pages, against 16 percent without one. That is not a rounding error. It is roughly half the click, gone.
Wang, Gleason, Bart, Wilson, and Metaxa ran a preregistered field experiment with 1,100 people. An AI Mode-only Google experience cut publisher click-through by 18.8 percentage points. Xu, Iqbal, and Montgomery crawled 55,393 trending queries from March 13 to April 21, 2026. Overviews appeared on 13.7 percent of queries and on 64.7 percent of questions. Almost 30 percent of the domains cited in those Overviews were not on the first page of classic results. Agarwal and Sen, as those authors cite them, found a 38 percent drop in organic clicks and a 33 percent rise in zero-click searches.
The practical implication is ugly and simple. If the answer appears on the results page, most people will not visit you. The work is to be the source inside the answer, not only to rank under it.
Chapekis, Lieb, Shah, and Smith, 2026.
| Behavior | Share of visits (%) |
|---|---|
| Click a result, no Overview | 15 |
| Click a result, with Overview | 8 |
| Click an Overview citation | 1 |
Two Markets, Two Sizes
Gartner and Forrester still have not published a GEO/AEO services TAM. Two commercial houses have. We would not wait for a third.
QY Research puts AEO tools and services at about $0.18 billion in 2023, $0.53 billion in 2024, and $1.11 billion in 2025. It projects about $9.1 billion in 2031, a 42 percent annual rate from 2025. In its 2025 split, generative-AI AEO is about 79 percent of revenue, and large enterprises about 64 percent of spend.
MarketsandMarkets uses a tighter definition of generative engine optimization: $390 million in 2025, reaching $4.25 billion in 2032 at 41 percent a year from 2026. For 2025 it splits the market $164 million in North America, $101 million in Europe, $86 million in Asia Pacific, and $39 million in the rest of the world.
SEO services dwarf both. Mordor Intelligence estimates $74.9 billion in 2025, $83.98 billion in 2026, and $148.86 billion in 2031, growing 12.12 percent a year. Even QY's 2031 AEO number is still a sliver of that.
The two firms are not describing the same basket. We would use them as a range, not pick a winner. Either way this is still small next to SEO services, which is why we think GEO/AEO spend arrives as a line inside SEO retainers before it becomes its own category at most firms.
QY Research (AEO) and Mordor Intelligence (SEO services), USD billions. Different definitions; log scale.
| Year | QY AEO (USD bn) | Mordor SEO (USD bn) |
|---|---|---|
| 2023 | 0.18 | |
| 2024 | 0.53 | |
| 2025 | 1.11 | 74.9 |
| 2026 | 83.98 | |
| 2031 | 9.11 | 148.86 |
2026-2031. QY publishes 2031. MarketsandMarkets publishes 2032 and a 41 percent CAGR. Years in between follow those rates.
| Year | QY Research AEO path (USD bn) | MarketsandMarkets GEO, implied (USD bn) |
|---|---|---|
| 2025 | 1.11 | 0.39 |
| 2026 | 1.58 | 0.55 |
| 2027 | 2.24 | 0.78 |
| 2028 | 3.18 | 1.09 |
| 2029 | 4.52 | 1.54 |
| 2030 | 6.41 | 2.17 |
| 2031 | 9.11 | 3.06 |
The GEO/AEO Market: A Forecast For The Next Five Years
This section is our forecast for the GEO/AEO market over the next five years, 2026 through 2031.
Our base case is that service and tool spend grows at something close to the 40 percent a year both QY Research and MarketsandMarkets published. QY's AEO path is 42 percent a year from 2025. The MarketsandMarkets GEO path is 41 percent a year from 2026. That is much faster than SEO services, which Mordor Intelligence puts at 12.12 percent a year. We do not think GEO replaces SEO in this window. We think it rides on top of it. A mid-market firm that already spends on SEO will start seeing an AEO or GEO task in the same statement of work. It will not see a separate $9 billion industry land on its accounts-payable desk.
If the narrower MarketsandMarkets definition holds, the 2031 market is on the order of $3 billion. If QY's wider AEO tools-and-services definition holds, more like $9 billion. We would be surprised by either a stall back to SEO-like growth or a jump past QY's path. The demand Pew and the click papers describe is already here. Measurement is still too sloppy to unlock enterprise-wide budgets overnight.
What companies should expect is more RFP language about ChatGPT, Gemini, and AI Overviews. More retainers that mix classic SEO with citation work. Tools that count mentions in answers. And a lot of vendors relabeling SEO as GEO without a method. The scarce skill is getting cited, not producing more pages.
What will still be annoying in 2031 is the same mess we already see. Engines will not cite the same pages. Xu, Iqbal, and Montgomery already find that almost 30 percent of Overview sources were not on the first page of classic results. Attribution from citation to revenue will still be argued in meetings. Google will still send most of the clicks.
If you sell into this market, the next five years look like a land grab in a small, fast overlay. If you buy, it looks like paying your SEO shop to care about answers, and checking whether they can show citations rather than rankings. We would not bet against the overlay. We would not bet that it becomes its own industry on most invoices.
References
- Agarwal, Saharsh, and Ananya Sen. 2026. "The Impact of Google AI Overviews on Publisher Traffic and User Experience: Evidence from a Field Experiment." Working paper, June 17, 2026. As cited in Xu, Iqbal, and Montgomery 2026. https://www.tse-fr.eu/sites/default/files/TSE/documents/sem2026/eco_platforms/agarwal_google_search_aio.pdf.
- Aggarwal, Pranjal, Vishvak Murahari, Tanmay Rajpurohit, Ashwin Kalyan, Karthik Narasimhan, and Ameet Deshpande. 2024. "GEO: Generative Engine Optimization." In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 5-16. New York: ACM. https://doi.org/10.1145/3637528.3671900.
- Bick, Alexander, Adam Blandin, and David J. Deming. 2026. "The Rapid Adoption of Generative AI." Management Science. As cited in Stanford Institute for Human-Centered AI 2026.
- Chapekis, Athena, Anna Lieb, Sono Shah, and Aaron Smith. 2026. "Investigating Click Behaviors On Google Search Result Pages That Produce an AI Overview." arXiv:2608.04831. https://doi.org/10.48550/arXiv.2608.04831.
- MarketsandMarkets. 2026. "Generative Engine Optimization (GEO) Market." Report code UC-TC-1141. https://www.marketsandmarkets.com/Market-Reports/generative-engine-optimization-geo-market-265868816.html.
- Mordor Intelligence. 2026. "Search Engine Optimization (SEO) Services Market Size." https://www.mordorintelligence.com/industry-reports/seo-market.
- Pew Research Center. 2026. "Americans and AI 2026: Chatbots, Smart Devices and Views on Impact." June 17, 2026. https://www.pewresearch.org/internet/2026/06/17/americans-and-ai-2026-chatbots-smart-devices-and-views-on-impact/.
- QY Research. 2025/2026. "Global Answer Engine Optimization (AEO) Service Market Outlook, Forecast to 2031." https://www.qyresearch.com/reports/5537157/answer-engine-optimization--aeo--service.
- Stanford Institute for Human-Centered AI. 2026. "Artificial Intelligence Index Report 2026." https://doi.org/10.48550/arxiv.2606.15708 and https://hai.stanford.edu/assets/files/ai_index_report_2026.pdf.
- Vishwakarma, Rahul, Shushant Kumar, and Ratnesh Jamidar. 2026. "What Gets Cited: Competitive GEO in AI Answer Engines." In Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '26), July 20-24, 2026, Melbourne, VIC, Australia. ACM. https://doi.org/10.1145/3805712.3808445.
- Wang, Stephanie T., Jeffrey Gleason, Yakov Bart, Christo Wilson, and Danaé Metaxa. 2026. "AI in Search Reduces Publisher Referrals Without Improving User Experience: Experimental Evidence." arXiv:2608.18352. https://arxiv.org/abs/2608.18352.
- Xu, Haofei, Umar Iqbal, and Jacob M. Montgomery. 2026. "Measuring Google AI Overviews: Activation, Source Quality, Claim Fidelity, and Publisher Impact." arXiv:2605.14021. https://arxiv.org/abs/2605.14021.