Seer Interactive analyzed 3,119 informational queries across 42 organizations between June 2024 and September 2025. Organic click-through rate on queries that showed an AI Overview fell from 1.76 percent to 0.61 percent, a 61 percent drop (Seer Interactive). Paid CTR on the same queries fell 68 percent. That is the number that reopened every CMO's budget conversation in 2025 and the number you have to start with when you walk into a 2026 planning meeting. Soar is a community marketing agency that has run 4,200+ community campaigns across 280+ brands since 2017, and over the last three years the question we get most often from marketing leaders is not "should we do AI visibility?" It is "where does the money come from."
The CTR cliff that changed the question
Before September 2024, "AI visibility" was a side project. After September 2025, it is a replacement. The Seer data is the clearest single number: a 61 percent organic CTR drop on AI Overview queries, and a 41 percent CTR drop on queries without AI Overviews, the latter driven by broader user behavior change, not AI Overview exposure (Search Engine Land). In plain terms: even the queries Google has not touched with AI are sending fewer clicks. Users are answering in ChatGPT, Perplexity, Gemini, or trusted brands. Google is still the volume leader, but it is no longer the only funnel.
The CTR number reframes the budget question. If SEO is delivering 40 percent fewer clicks per ranking position than it did 18 months ago, an SEO retainer at the same price is producing fewer outcomes. Either you accept the degraded output, or you reallocate. Most brands that do not reallocate are quietly running a 30 to 50 percent real-terms cut to their organic acquisition.
What AI visibility is actually replacing
"AI visibility" gets used to mean three different things, which confuses budget conversations. It is worth being precise about what you are buying:
- AI citations. Being cited by name in ChatGPT, Claude, Perplexity, or Google AI Overviews when users ask category or comparison questions. This is what most GEO work targets (Aggarwal et al.).
- AI Overview inclusion. Showing up inside Google's AI Overview panel on the SERP (Google Search AI features). This is closer to classic SEO because the surface is still Google.
- AI-assistant product placement. Being recommended when ChatGPT Search or similar agent-style tools answer "best X for Y" queries (ChatGPT Search).
The three have different source pools. AI Overview inclusion leans heavily on Google's index. AI citations across ChatGPT, Claude, and Perplexity lean on Reddit, Wikipedia, YouTube, and a long tail of trusted editorial and review sites (Semrush most-cited domains). AI-assistant product placement leans on comparison articles, review sites, and structured data.
That matters for budgeting because the work products are different. AI Overview inclusion is mostly SEO hygiene. AI citations require community presence and external source placement. AI-assistant product placement requires comparison content and review-site relationships. An agency pitching "AI visibility" as a single service needs to tell you which of the three they are buying you.
The budget allocation framework for 2026
The framework below is what we use when marketing leaders ask us how to split an existing organic budget. The starting point matters: a brand with no SEO program should not copy the allocation used by a brand with a $40K/month SEO retainer. The table reads as a reallocation, not a total budget recommendation.
| Brand profile | SEO | AI Overview inclusion | AI citations (Reddit/editorial) | AI-assistant placement |
|---|---|---|---|---|
| Mature SEO program, slow-changing category | 55% | 15% | 20% | 10% |
| Mature SEO program, fast-changing category | 40% | 15% | 30% | 15% |
| Early-stage, little SEO equity | 25% | 10% | 45% | 20% |
| B2B SaaS, high comparison volume | 35% | 10% | 30% | 25% |
| Consumer brand, high review volume | 40% | 15% | 30% | 15% |
A few notes on how these numbers actually play in planning. First, the "mature SEO, slow-changing category" row is the one most brands mistake themselves as. Most B2B SaaS categories move fast enough that the second row is closer to reality. Second, the "early-stage" row pulls hard toward AI citations because Reddit, editorial, and community content compound faster for a young brand than an SEO program that needs 12 to 18 months to produce rankings. Third, AI Overview inclusion percentages are low because the work is largely covered by good SEO hygiene, not separate spend, it is less a line item and more a discipline applied to existing content.
For a quantitative starting point: a brand running a $20,000/month SEO retainer in 2024 is, in most scenarios, overbudgeted on SEO for 2026. The honest adjustment is often a reallocation of $4,000 to $6,000 per month into AI visibility work, not a net-new $6,000 budget line. CMOs who frame it as reallocation close the budget approval in one conversation. CMOs who frame it as net-new spend spend a quarter fighting for it.
Where the two disciplines overlap (and where they don't)
Every SEO agency pitching AI visibility will tell you the two disciplines overlap. They do, just not as much as the pitch implies. The overlap is real in content structure and entity hygiene. It breaks down in source strategy and measurement.
| Area | SEO | AI visibility | Overlap? |
|---|---|---|---|
| Content structure | H2 hierarchy, answer capsules, schema | Same, plus question-style H2s and FAQs | High |
| Link sources | DR-weighted backlinks | Citation sources (Reddit, YouTube, Wikipedia, editorial) | Low |
| Technical SEO | Crawl, indexation, Core Web Vitals | LLM crawlability (GPTBot, ClaudeBot, PerplexityBot) | Medium |
| Measurement | Rankings, clicks, impressions | Share of voice across LLMs, citation frequency, prompt-level rank | Low |
| Entity authority | Brand mentions, knowledge panels | Same plus structured brand consistency across sources | High |
| Community presence | Not a core input | Core input (Reddit is the single largest LLM source) | None |
The "none" row is the one to focus on. Community presence, specifically Reddit, Quora, and category-specific forums, is the single largest input to AI citations, and it sits entirely outside the classic SEO playbook. An SEO agency that claims to do AI visibility without a Reddit or community capability is, in practice, doing AI Overview inclusion and hoping the rest shows up. It does not show up on its own.
Measurement: why AI visibility breaks SEO dashboards
SEO dashboards are built around a single surface: the SERP. Rankings, clicks, impressions, CTR. AI visibility is measured across multiple engines with different retrieval logic, different citation formats, and different update cadences. ChatGPT cites differently from Claude, which cites differently from Perplexity, which cites differently from Google AI Overviews. A single "share of voice" number hides as much as it reveals.
The measurement stack for AI visibility is closer to a brand tracker than an SEO report. You are tracking prompt-level inclusion across a target prompt set, citation frequency by engine, share of voice against a defined competitor set, and sentiment inside those citations. None of that shows up in Search Console. Most SEO agencies adding AI visibility as a bolt-on use the same dashboard they already had and quietly drop a "ChatGPT mention rate" number on top. That is directional, not diagnostic. Our AI visibility measurement framework walks through the stack that actually works.
For budget purposes: plan for a separate measurement tool or dedicated analyst cost inside the AI visibility line. $500 to $2,000 per month for tooling is typical in 2026. Free tools exist but require an analyst to pull them together weekly.
Team skills: can your SEO team do this
The honest answer is: partially. An SEO team with a strong content lead, a technical SEO, and a link-building function can cover AI Overview inclusion and part of AI-assistant placement without new hires. The two capabilities that almost never transfer are community operations and editorial placement. Both require outside specialists or dedicated internal headcount.
The practical move for mid-market brands is usually one of three options:
- Retain SEO, add a community-focused AI visibility agency for Reddit, Quora, and editorial. This is our most common engagement structure.
- Hire a dedicated AI visibility lead internally (usually a senior content strategist with community experience). Viable above a $50M company with 3+ marketing headcount.
- Consolidate with one agency that covers both. Rare, because the capability sets are genuinely different. If pursued, the test is whether the agency has named community operators on staff, not whether it has a deck slide on GEO.
A conditional recommendation: three scenarios
Scenario A: Mid-market B2B SaaS, $20K/month existing SEO retainer. Reallocate $4,000 to $6,000/month from SEO to AI visibility. Split the new budget: 60 percent to community and editorial placement, 20 percent to comparison content, 20 percent to measurement and tooling. Expect first measurable citation lift in 90 to 120 days.
Scenario B: Consumer brand, $10K/month SEO retainer. Keep SEO flat, add a $5,000 to $8,000/month AI visibility program on top. Budget justification: 70 percent of review and "best-of" queries now route through AI before the user lands on a SERP. Not showing up in those paths costs more than the incremental spend. Expect first Reddit citations in 6 to 8 weeks; first AI assistant recommendations in 90 to 150 days.
Scenario C: Early-stage startup, no SEO program. Skip the SEO-first build. Spend the first 9 months on community presence (Reddit, Quora, Hacker News, category forums) and editorial placement. Layer SEO in month 10 once the brand has category presence to link from. The right starting budget is $8,000 to $12,000/month for community plus modest content investment, and the 18-month payback beats most alternatives.
FAQ
Should I stop doing SEO entirely? No. Google still drives the majority of informational search volume in 2026 (First Page Sage), and AI Overview inclusion is won by strong SEO fundamentals. Cut the lowest-performing SEO content, not the program.
How long before AI visibility produces measurable results? First citations typically appear in 60 to 120 days with consistent community and editorial work. Meaningful share-of-voice change takes 6 to 9 months. Anything faster is either a small sample (one or two prompts) or an already-authoritative brand.
Is AI Overview optimization the same as GEO? No. GEO is the broader discipline of optimizing for generative engines, which includes ChatGPT, Claude, Perplexity, and Google's AI Overviews (eMarketer). AI Overview optimization is a subset, it covers only the Google surface. Most brands need both.
What tools do I actually need? At minimum, a prompt-tracking tool (Profound, Peec, or similar), a Reddit monitoring stack, and a traditional SEO tool you already pay for (Ahrefs or Semrush). Budget $1,000 to $2,500/month for the AI-specific tooling in 2026. Our list of free tools covers the starter stack.
Will my PR team help or hurt this? Help, if they understand that tier-1 editorial mentions are AI citation fuel. Most PR teams think in terms of coverage volume. Tell them you want mentions in the sources ChatGPT and Claude actually cite, and the story list tightens fast.
If your 2026 plan is still mostly SEO and you are trying to decide where to reallocate, we can help you model scenarios and map the community and editorial work to your category. Request a proposal and we will send back a scoped plan with the citation targets we think are reachable in 90 and 180 days.