Do GEO Tactics Stop Working When Everyone Uses Them?
Two simulations of GEO in a competitive market: fixed tactics lose their edge as adoption rises, and the ranking signal itself drifts away from quality as the pool optimizes.
Updated · Published
Frequently asked questions
Do GEO tactics stop working when competitors use them too?
In simulation, yes — and past a threshold they go negative. A 2026 preprint from Capital One AI Foundations tested fixed rewriting tactics while varying the share of competing documents also optimized, and found that "single-strategy methods at full competitive optimization (α=0.8) drop below the unoptimized (α=0) baseline." A method that re-selected tactic combinations based on the competitive context lost only 11.4% of its performance across the same range.
Which GEO tactic works best?
No single tactic won. On GEO-Bench, the strongest single tactic tested (statistics addition, at 23.08 PAWC) trailed both an agentic rewriting baseline (27.95) and the paper's combination-search method (32.62). The consistent finding across the KDD 2024 GEO paper and this one is that combinations beat single edits, and this paper adds that the best combination depends on what competitors have already done.
Is the claim that GEO boosts visibility by 40% still valid?
As a measurement of one document being optimized while competitors stand still, yes — that is what the KDD 2024 paper measured. The 2026 competitor-aware study argues that this vacuum assumption is the limit of the finding: as the share of optimized competitors rises in its simulations, the advantage of any fixed tactic shrinks and eventually inverts. The 40% figure describes an early-adopter market, not a saturated one.
How close is the real web to the saturated scenario?
Not close, on average. The first measured GEO prevalence, from an August 2026 CISPA-led study, found 8.90% of pages retrieved for real queries were optimized, rising to 16.36% among pages modified in 2026 — against the 80% adoption where simulated single tactics went negative. But prevalence varied from 0% on major reference and medical sites to 20% on sampled Amazon pages, so a competitive commercial category can be far ahead of the average.
What is competitor-aware GEO?
Competitor-aware GEO selects rewriting strategies based on the competitive context — which documents share the retrieval pool and how optimized they already are — instead of applying a fixed checklist. The paper that names it searched 32,768 possible combinations of 15 published tactics per query using Bayesian optimization, then fine-tuned a model to propose combinations directly. It beat the strongest agentic baseline by 4.67 PAWC on GEO-Bench and by more than 18% on the competitive variant.
What happens to content quality when everyone optimizes repeatedly?
In the CHASE simulation, posted 31 August 2026 and accepted to COLM 2026, 20 rounds of creators adapting documents to the same LLM ranking signal reduced the alignment between ranking and independently judged quality in all six domains tested, with Spearman's ρ falling by between 0.018 and 0.107 (mean 0.068). A random-target control showed the decline comes from chasing the ranking signal specifically, not from repeated rewriting itself. Alignment stayed positive at round 20 and measured homogenization was modest — erosion, not collapse.
Does ranking well actually predict getting cited?
Closely enough to optimize against, per the CHASE validation: ranking position predicted whether a document was cited in grounded generated answers with an AUC of 0.853 ± 0.093 across six domains. That matters for reading GEO research, because it means results measured on ranking proxies transfer reasonably well to the citation outcomes practitioners actually track.
Does optimizing for AI citations make content less accurate?
In this study's simulations, the visibility-optimal rewrites were measurably less faithful to the source: 4.90 versus 8.29 for unrewritten documents on an LLM-judged scale. That matches the failure mode in a separate 2026 mechanism-design paper, where repeated optimization rounds accumulated unsupported claims. Visibility and fidelity are separate objectives, and optimizing hard for one traded away the other in both papers.
Should I stop running the GEO checklist?
No — the measured web is still mostly unoptimized, and in a low-adoption category the published tactics remain the best-evidenced starting point. What the research removes is the assumption that the checklist's effectiveness is a property of the checklist. It is a property of your category's adoption rate, which means it needs to be measured in your category, and re-measured as the category fills in.
Every ranking of GEO tactics you have read shares an assumption: your competitors' pages hold still while you optimize yours. A preprint posted on 27 August 2026 by Vaibhav Sourirajan and colleagues at Capital One AI Foundations drops that assumption and re-runs the question. When the share of competing documents that are also optimized rises to 80%, every fixed single tactic they tested scores below the baseline of never having optimized at all.
0% on Wikipedia and major medical sites, 20% on sampled Amazon pages
Adoption trend by year of modification
2024 7% → 2025 12.8% → 2026 16.4%
Faithfulness cost of the winning rewrites
4.90 against 8.29 for the unrewritten documents
CHASE: quality–ranking alignment over 20 rounds
mean Δρ −0.068 across six domains, control ≈ 0 — arXiv:2608.30466
Key takeaways
Tactic effectiveness is relative, not absolute. The paper formalizes GEO as a competitor-aware selection problem because "as adoption of content optimization grows, optimal strategies for rewriting content change."
Combinations beat checklists. Searching the 32,768 possible combinations of 15 published tactics per query scored 32.62 PAWC on GEO-Bench, against 27.95 for the strongest agentic baseline and 23.08 for the best single tactic tested.
Saturation inverts the sign. At 80% simulated adoption, fixed single-strategy methods "drop below the unoptimized (α=0) baseline." The competitor-aware method lost only 11.4% across the full adoption range by shifting "toward less saturated alternatives."
The real web is nowhere near 80% — yet. Measured prevalence is 8.90% overall and 16.36% for 2026-modified pages, but it ranges from 0% to 20% by domain, so the number that matters is your category's.
Added 1 September: a second simulation, CHASE, generalizes the result to the whole pool. Over 20 rounds of creators adapting to one LLM ranking signal, the correlation between rank and independently judged quality fell in all six domains tested (mean Δρ −0.068) — the tactics stop paying you, and the signal gets worse at rewarding quality at the same time.
The trade the paper underplays: its winning method's rewrites scored 4.90 on faithfulness against 8.29 for the originals. The most visible version of your page was not the most accurate one.
The usual caveats apply, and then some: simulated competitors, proxy engines, a word-count metric, and a method that reads the entire competitive corpus before choosing — an advantage no publisher has.
Updated 11 September 2026: added key facts and the CHASE results table. The CHASE simulation itself was added on 1 September 2026.
Since the KDD 2024 GEO paper reported that optimization "can boost visibility by up to 40%," the field has treated its tactic list — add statistics, cite sources, add quotations, improve fluency — as a ranked menu. The ranking came from a specific experimental shape: one document in a retrieval pool gets rewritten, the others stay as they were, and the change in the engine's answer is measured.
That was a reasonable design in 2023, when almost nobody was optimizing. It describes a land grab. The Capital One paper asks what the same tactics are worth once the land is taken, and its answer is that the ranking itself is unstable: the value of any edit depends on how many competing documents already carry it.
The intuition is familiar from every prior optimization discipline. If one page in five has extractable statistics and yours is it, the engine's preference for statistics is your advantage. If four in five do, the same edit buys parity at best. What the paper adds is machinery for measuring this and a demonstration of how sharp the falloff gets.
The method has two phases. First, for each query and document set, it runs Bayesian Optimization of Combinatorial Structures — BOCS — over the 2^15 = 32,768 possible on/off combinations of 15 rewriting strategies drawn from the published GEO literature, using 50 evaluations per search. The 15 include the KDD 2024 tactics (statistics addition, citing sources, quotation addition, fluency, authoritative tone, keyword stuffing) alongside later additions like structured formatting and LLM guidance from C-SEO Bench.
Second, because running BOCS per page is impractical, the observations from those searches become training data: preference pairs and reasoning traces that fine-tune a language model to look at a document corpus and propose a strategy combination directly.
Visibility is scored with PAWC — Position-Adjusted Word Count — the standard GEO-research metric that credits a document for the words the answer draws from it, discounted by position. On GEO-Bench with gpt-oss-120b as the engine:
Method
PAWC
Competitor-aware selection (this paper)
32.62 ± 0.25
AgenticGEO (strongest baseline)
27.95 ± 0.30
Statistics addition (strongest single tactic)
23.08 ± 0.28
Fluency optimization
21.92 ± 0.32
The gap worth noticing is not the headline one. It is that the best single tactic — the one at the top of most practitioner checklists — sits five points below an agentic rewriter and nearly ten below combination search. The checklist's unit of action, the individual tactic, is the wrong granularity even before competition enters.
Then competition enters. The authors build geo-bench_comp, a variant in which a fraction α — the share of competing documents that are themselves GEO-optimized — is rewritten by simulated optimizers, and test α from 0 to 0.8 — from an unoptimized field to one where four of five competitors are running GEO.
Two results carry the paper. Averaged across adoption rates, the competitor-aware method scores 29.93 PAWC against 25.20 for the strongest baseline — an 18% edge. And at the high end, the fixed approaches do not merely lag: "single-strategy methods at full competitive optimization (α=0.8) drop below the unoptimized (α=0) baseline." Running the standard playbook in a saturated category simulated as worse than never having run it.
The competitor-aware selector avoids this by doing the thing its name says: it infers the adoption level from the documents it can see and shifts "toward less saturated alternatives as the competitive landscape is increasingly optimized." Its total degradation across the full adoption range is 11.4% — it bends where the fixed tactics break. The method also transferred out of distribution, scoring 37.78 on an e-commerce set and 33.48 on Researchy-GEO, more than 8% above the strongest baseline on each.
For anyone doing answer engine optimization, the mechanism matters more than the scores. Nothing about a statistics block or an FAQ became less legible to the engine at high α. What changed is that the edit stopped being a difference. Engines cite what stands out from the pool; an optimization everyone applies is, by definition, no longer what stands out.
The paper does not say, because it cannot — its competitors are synthetic. But the measurement exists elsewhere, and reconciling the two is where this becomes usable.
The CISPA-led GEO-Flag study, posted ten days earlier, put the first number on real-world adoption: 8.90% of pages retrieved for 1,000 real queries were GEO-optimized, rising to 16.36% among pages modified in 2026 — we covered it in our analysis of GEO prevalence. Against the simulation's axis, the measured web sits near α = 0.1, one page in six for recent content. The regime where fixed tactics go negative starts around α = 0.8. On the web average, the 2024 playbook is still operating in something like the vacuum it was measured in.
But the average is doing heavy lifting. The same study found 0% detected optimization on Wikipedia and major medical sites and 20% on sampled Amazon pages, and its year-over-year trend — 7% of 2024-modified pages, 12.8% for 2025, 16.4% for 2026 — points one direction. A competitive commercial category is plausibly years ahead of the mean, and it is precisely the categories where AI answers drive purchases that fill fastest.
That yields a decision rule the paper stops short of stating. The question "do GEO tactics work" has no fixed answer; the question "how optimized are the pages cited for my prompts" does, and it determines the first one. Pull the citations for your tracked prompts, assess how many are visibly running the playbook, and read your position off the curve: a quiet pool means the standard tactics still buy differentiation, a crowded one means they buy parity, and the trend line tells you how long the first state lasts.
Added 1 September 2026. The Capital One paper asks what saturation does to your tactic. A second simulation posted on 31 August — CHASE, by Qianwen Gao and colleagues, accepted to COLM 2026 — asks what it does to everyone: it runs 20 rounds in which creators repeatedly adapt documents toward features extracted from recent ranking winners, across six domains (retail, video games, books, web, news, debate), with roughly 29% of non-winning documents rewritten each round.
Before running it, the authors validate the thing most GEO research quietly assumes: that ranking is a fair proxy for citation. In their measurement, a document's rank predicted whether it was cited in grounded generated answers with an AUC of 0.853 ± 0.093 across the six domains. That number is useful on its own — it says rank-shaped optimization and citation-shaped optimization are close to the same game, which is why results like the adoption curve above transfer to the citation outcomes brands actually track.
The core result: quality–ranking alignment declined in every domain, with the change in Spearman's ρ ranging from −0.018 to −0.107, mean −0.068 over the 20 rounds.
Change in quality–ranking alignment (Δρ) over 20 rounds
Mean across six domains
−0.068
Best case (books)
−0.018
Worst case
−0.107
Retail, ranking-derived rewrite targets
−0.047
Retail, random rewrite targets (control)
+0.001
The control condition is what makes it attributable: documents rewritten just as often, but toward arbitrary targets rather than ranking-derived ones, showed little or none of the decline. The drift comes from chasing the signal, not from rewriting per se.
Two honest readings the paper's title undersells. Alignment eroded but stayed positive at round 20 in every domain — after 20 rounds of exploitation the ranker still weakly favored better documents. And the homogenization in the acronym is measurable but small: mean pairwise embedding similarity within a query pool moved by a point or two (video games went from .334 to .348), and the retail pool actually got slightly more diverse. This is a direction of travel, not a collapse — the same reading discipline the adoption curve deserves.
Put the two papers together and the saturation story gains a second axis. As your category's α rises, the Capital One result says any fixed tactic buys less; CHASE says the referee's ordering drifts away from quality at the same time. The compounding conclusion for a brand is the one this post already reached from the faithfulness table: differentiation that survives is the kind no rewrite reproduces, because rewriting toward the signal is precisely the behavior that degrades the signal.
The paper logs this as a trade-off in a table. For a brand it is the whole risk, and it deserves more than a footnote.
The winning method's rewrites scored 4.90 on faithfulness against 8.29 for the unrewritten documents on the paper's LLM-judged scale. The visibility-optimal version of a page drifted furthest from what the page originally said. A rewrite that wins the citation by saying something your source material does not support becomes an engine confidently attributing that claim to you — and unlike a ranking, you do not get to see it happen.
Four independent groups now converge on this. The mechanism-design work on citation wars found unsupported claims accumulating over repeated optimization rounds. The GEO-Flag audit found decoration-grade citations on 69% of the pages it detected as optimized. This paper measures the faithfulness cost directly. And the CHASE result above is the population-level rhyme: the same pressure decouples the ranking signal from quality for everyone at once. Pushed hard enough, visibility optimization manufactures unsupported claims.
The operational consequence is the one thing on the Monday list that is not a measurement task: read every optimized rewrite against its source, and treat a claim that appears only in the optimized version as a claim you did not make until the optimizer made it for you.
The competitors are simulated — three optimizer functions applied on a fixed rotation — and the engines are proxies: gpt-oss-120b and Llama-3.3-70B-Instruct, not the production systems answering buyers' questions. PAWC measures words drawn into an answer, which is an impression proxy, not a mention, a recommendation, or a sale. And the competitor-aware method holds what the authors concede is "an information advantage": it reads the entire competitive corpus before choosing a strategy. No publisher sees the retrieval pool the way this selector does — though tracking which pages engines actually cite for your prompts is the closest available approximation.
The CHASE simulation carries its own caveats in the same family: 20 queries per domain with 5–10 candidate documents each, an LLM ranker (Gemini Flash-class) standing in for a production engine, and quality judged by another LLM whose agreement with human raters was moderate (Spearman's ρ of 0.58 on shared dimensions). The authors are explicit that the framework "intentionally isolates repeated adaptation to an LLM ranking signal rather than modeling a complete generative-search system."
There is also a quieter irony worth holding: a paper about competitors adapting to each other is itself now public. A tactic selector that wins by finding "less saturated alternatives" describes an equilibrium that moves again the moment enough parties run one — and parties will: a demo system automating exactly this kind of feedback-driven tactic selection, Agent2UCB, was posted two days later.
Establish your category's adoption rate. Take the domains cited for your tracked prompts and assess the pages: statistics blocks, FAQ scaffolding, citation-dense passages, the recognizable shape of the playbook. That number — not the 9% web average — is your α.
Stop reading the 2024 tactic ranking as a priority order. Single tactics were the weakest class of method even without competition. Test combinations against your own prompt set and keep what moves your numbers.
Put your differentiation in what cannot saturate. Every tactic in the 15-strategy space is available to your competitors the day they want it. First-party data, original numbers, and named expertise are the additions no rewriting strategy reproduces.
Read every optimized rewrite against its source. The measured faithfulness cost means the burden of proof sits on the rewrite. A claim that appears only in the optimized version is a claim you did not make until the optimizer made it for you.
Timestamp your conclusions. Whatever you measure about tactic effectiveness this quarter describes this quarter's adoption level. The GEO-Flag trend line says the pool changes underneath you; our own tracking shows 60–70% day-to-day turnover in cited domains even now.
The honest summary is that GEO tactics work the way arbitrage works: the return is real, and it is a function of how many others have found the trade. The KDD 2024 measurements described an empty field, the CISPA measurements say the field is one-sixth full and filling, the Capital One simulation shows the far end where the standard moves pay less than nothing, and CHASE shows the trade eroding the price signal itself as the field fills. The practitioner's job shifts accordingly — from running a checklist to knowing where on that curve their category sits, which is a measurement question before it is a tactics question.
Elmo is an open-source, self-hosted AI visibility platform that runs your prompt sets across ChatGPT, Claude, Gemini, Perplexity, and Google's AI surfaces and records every answer with its citations and competitor mentions. Because the cited pages for each prompt are in your own database, the adoption-rate check this research calls for is a query, not a project — and re-running it each month tells you when your category's curve starts to bend.