What This Data Measures — and What It Doesn't
The figures in this report come from Google's Keyword Planner, the research tool built into Google Ads. It tracks how often specific phrases are typed into Google Search — and because it comes from Google itself, it is the closest thing available to a primary source on search behavior.
The purpose here is not to analyze a marketing tool. It's to use that tool as a window into a meaningful public behavior: how often people in the United States are actively searching Google for drug and alcohol treatment. When tens of thousands of people type "rehab centers near me" into Google in a given month, that is a signal about something real — awareness, need, urgency, help-seeking. When those numbers fall significantly year over year, that also means something, even if the exact reason isn't always clear from the data alone.
Google also groups keywords with similar meanings together, which means different search phrases can show identical volumes not because they behave identically, but because Google has placed them in the same reporting cluster. Several places in this data where five or six unrelated terms show the exact same number are almost certainly this grouping effect at work rather than a coincidence.
One more honest caveat: Google has historically adjusted its bucket structure and grouping logic over time, typically without announcing it. These changes tend to surface when researchers notice anomalies in the data — identical percentage shifts across unrelated terms, or volumes that seem to jump between report pulls without clear explanation. We have no specific evidence of such a change during the April 2025 to April 2026 window covered here. But the possibility cannot be ruled out, and if Google did reclassify how certain terms are grouped during this period, some portion of what appears as a volume decline could reflect a data methodology change rather than a change in actual search behavior. This is an inherent and unresolvable limitation of the tool.
Finally, Keyword Planner has no visibility into AI-driven search surfaces — queries handled inside Google's AI Mode, ChatGPT, Perplexity, or other platforms leave no trace in this data. That matters more for some query types than others, and we address it directly in the AI section later in this report.
Sharp, Broad Declines — With a Narrow Band of Exceptions
Across 396 substance abuse and addiction treatment keywords tracked from April 2025 through April 2026, the direction is unmistakably down. Search volume declined across the majority of terms in every single category — alcohol treatment, drug treatment, detox, outpatient programs, and general rehab searches alike. This is not a story of a few weak terms pulling down an average. The declines are broad, consistent, and in many cases severe.
The top-line 18% decline in the largest general search terms — phrases like "rehab centers near me" — represents thousands of fewer monthly searches and is the floor of this story, not the ceiling. Alcohol-specific terms fell far harder: the six highest-volume alcohol treatment keywords declined an average of 54% year over year. Mid-tier drug treatment terms fell 45–56%. The pattern is consistent — the more specific the term, the steeper the decline.
A narrow band of terms did grow. Some recovery-framed language showed genuine gains, and a handful of terms posted sharp 3-month spikes heading into April 2026. These exceptions are noted where relevant. But they represent a small minority of terms, most at relatively modest search volumes, and they do not change the overall direction of the data.
What we cannot say from this data is why. The data records what people typed. It doesn't tell us whether fewer people needed help, whether they found it another way, or whether something is disrupting how people search. Those questions are addressed directly in the AI section — and in the mental health parallel that follows it.
Alcohol-Specific Treatment Searches: The Steepest Declines in the Dataset
Of all the categories in this report, alcohol-specific treatment searches show the largest and most consistent drops year over year. This isn't a case of a few outlier terms dragging down an average — 33 out of 36 tracked alcohol keywords declined, and the declines are deep across the board.
All major alcohol treatment terms declined consistently across the year. "Alcoholic center for women" is the lone exception, climbing steadily to reach its highest point in April 2026.
The full year-over-year picture
| Keyword | Apr 2025 | Apr 2026 | Yr/Yr | 3-mo trend |
|---|
What the monthly data shows
Looking month by month, the decline in alcohol treatment searches wasn't a single drop — it was gradual and sustained across the year. Most terms were already down significantly by summer 2025 compared to spring. The trough came around November and December 2025, after which several terms partially recovered through the first quarter of 2026. But that recovery brought most terms back only to levels still well below where they were in April 2025. The three-month trend figures heading into April 2026 are positive for several alcohol terms — "alcohol treatment programs near me" is up 23%, "drug and alcohol programs near me" up 24% — but those gains are measured off a deeply depressed base, and represent a partial bounce, not a return to prior levels.
Outpatient searches collapsed
Outpatient-specific alcohol terms tell a particularly sharp story. "Alcohol outpatient near me" registered 110 searches in April 2025, which is already a modest number. The monthly data shows it briefly spiked to 720 in September and October 2025 — then fell to just 30 by April 2026, ending lower than it started. "Alcohol outpatient program" declined 64% year over year. Whatever drove that mid-year spike didn't hold, and these terms have essentially disappeared from the data.
The one consistent bright spot
"Alcoholic center for women" grew from 480 searches in April 2025 to 720 in April 2026, up 50%, with a smooth upward trajectory through the year rather than a spike-and-drop pattern. It is the only alcohol treatment term in the dataset that shows consistent, sustained growth on both a year-over-year and recent basis. To put that in perspective: 720 monthly searches is a fraction of what the largest alcohol terms were registering even after their steep declines — "alcoholic rehab centers near me" alone still registers around 1,900 searches per month after falling 71%. The bright spot is real, but it operates at a volume that does not meaningfully offset the losses elsewhere in the category. "Alcohol addiction near me" also registered a positive year-over-year figure — up 191% — but this is driven by a single April 2026 spike to 320 after averaging 20–50 searches for most of the preceding year. Whether that holds is unknown.
Drug Treatment Searches: A More Mixed Picture
Drug-specific treatment searches show a different pattern than alcohol — the high-volume terms held up considerably better, while declines steepened significantly in the mid-range and lower-volume terms. The category as a whole is still predominantly down year over year, but it is not as uniformly or as deeply negative as alcohol.
The top drug treatment terms declined moderately. "Drug detox centers" showed a year-over-year gain, though the monthly data shows a dip back in April 2026 after a period of elevated volumes mid-year.
Year-over-year by keyword
| Keyword | Apr 2025 | Apr 2026 | Yr/Yr | 3-mo trend |
|---|
The five highest-volume drug treatment terms — "drug treatment centers near me," "drug rehabilitation center near me," "rehabs for drugs near me," "drug recovery near me," and "drug abuse treatment centers near me" — moved as a cluster from around 74,000 to around 60,500 year over year. The uniformity of that shift across five separate keyword strings is almost certainly Google's grouping behavior rather than five independent trends. The signal is that this cluster of broadly-worded drug treatment searches is down roughly 18% as a category.
Below that level, declines steepen considerably. "Drug rehabilitation facilities near me" fell 45%. "Drug detox programs near me" fell 56% — a term that had roughly 10,000 monthly searches in April 2025 and now registers around 4,400. "Drug and rehabilitation center" fell 46%. The pattern is consistent with alcohol searches: the more specific the term, the sharper the decline.
"Drug detox centers" registered a year-over-year gain, moving from around 12,100 to 14,800 — notable given the broader trend, though the monthly data shows the term dipped back in April 2026 after elevated volumes in early 2026, so it is not a clean upward story.
Broader Treatment Searches: Smaller Declines, but Still Down — With a Minority of Growing Terms
The broadest category in this dataset — terms like "rehab centers near me," "detox centers near me," "treatment centers near me," and "recovery center near me" — shows smaller year-over-year declines than alcohol or mid-tier drug searches, and contains most of the genuinely growing terms in the entire report. That is a relative statement, not an absolute one: the largest terms in this category still lost tens of thousands of monthly searches year over year, and the majority of terms here declined as well.
One thing worth acknowledging upfront: these terms don't exclusively capture substance abuse searches. Words like "rehab," "recovery," and "treatment" appear in searches related to physical rehabilitation, sports injury recovery, and medical treatment as well. There's no clean way to separate those out within this data. "Detox" skews more clearly toward substance use, but even there the word appears in general wellness contexts. This cross-category overlap likely provides some cushion in the volume figures — and means that reading these numbers as purely addiction-related requires some caution.
"Rehab centers near me" held in the 74,000–110,000 range throughout the year. "Recovery center near me" spiked sharply in April 2026, reaching 8,100 — its highest point in the dataset.
Year-over-year by keyword
| Keyword | Apr 2025 | Apr 2026 | Yr/Yr | 3-mo trend |
|---|
"Rehab centers near me" and "rehabilitations centers near me" each registered around 110,000 searches in April 2025 and 90,500 in April 2026 — down 18%, which represents tens of thousands of fewer monthly searches and should not be read as a mild outcome. "Detox centers near me" held flat at 27,100 in both April months. "Rehab centers" (without the location modifier) was flat year over year at 74,000 with a +49% 3-month trend, suggesting a meaningful recent uptick. "Treatment centers near me" was down 18% but showed a +23% 3-month trend heading into April 2026.
Several terms in this category are genuinely growing. "Recovery center near me" rose from 4,400 to 8,100, up 84%, with the April 2026 figure being the highest single month in the dataset and a 3-month trend of +125%. "Detox centers" (without near-me) grew 22%, from 12,100 to 14,800. "Residential rehabilitation centers" grew 48%. "Recovery facilities near me" grew 24%. These are real gains — but they are a small minority of terms in the category, and they do not offset the broader pattern of decline.
Luxury rehab searches dropped sharply
Within this broader category, the high-end and luxury subcategory stands out as the weakest segment by a large margin. "Luxury rehab centers," "luxury rehabilitation center," and "luxury rehab facilities" each fell from 8,100 to 2,400 — down 70% — with a −64% 3-month trend, meaning the decline is ongoing rather than stabilizing. "Luxury rehab Florida" fell from 3,600 to 720, an 80% drop year over year. Florida-specific luxury searches show the sharpest drops of any facility-type subset in the data.
How AI Search Fits Into This Picture
One question that comes up when interpreting any declining Google search data right now is: how much of this is people migrating to AI tools rather than people searching less overall?
AI-powered platforms have grown substantially. ChatGPT processed approximately 2.5 billion prompts per day as of mid-2025. By May 2026, according to traffic analysis by Ahrefs, ChatGPT's share of web referral traffic had surged 59.4% in a single month, while Google's share declined 5.7% in the same period. Google's share of web referral traffic had eroded from roughly 35% in mid-2025 to approximately 27% by May 2026.
Research by Datos and SparkToro found that Google searches per U.S. user fell approximately 20% year over year in 2025 — even as total information-discovery sessions grew — because more of that discovery was happening in AI sessions. The key distinction here is between search queries and website clicks: AI tools are absorbing a portion of the questions people used to bring to Google, which means fewer Google searches even when underlying demand for information hasn't necessarily changed. As Rand Fishkin of SparkToro described it:
"AI answers have dramatically altered the way many users engage with Google, answering their questions before they ever need to click on an organic result or perform a second, third, or fourth search."
This context matters — but the addiction and substance abuse treatment category sits in a different position than most search categories when it comes to AI.
Addiction searches are largely excluded from AI answers
Google has a consistent policy of excluding addiction queries from AI Overviews. Research by BrightEdge tracking Google's AI Overview deployment across three years confirmed that addiction queries have been "consistently excluded from AI Overviews across all three years." When someone searches for an alcohol rehab center or drug treatment program on Google, they are not receiving an AI-generated answer — they see a traditional search results page. For more explicit crisis-adjacent queries, Google routes users to the 988 Suicide and Crisis Lifeline rather than generating recommendations.
Other AI platforms apply their own caution. ChatGPT and similar models treat addiction treatment as a "Your Money or Your Life" category — meaning the platforms are more likely to refer users to professional resources than to provide specific facility names or recommendations. OpenAI stated in mid-2025 that its systems are "designed to identify distress, safely handle harmful requests, and guide users to real-world help." Claude, Anthropic's AI, is similarly cautious with direct treatment recommendations and avoids providing information that could facilitate harm in the substance use context. Perplexity is generally considered more permissive with factual health information and does surface treatment-related results — but its total query volume is a small fraction of Google's.
A 2025 study from the Recovery Research Institute examined specifically whether AI tools like ChatGPT can provide "safe and accurate answers to questions about substance use" — the fact that this was a live research question in 2025 reflects how unsettled this area still is. The platforms themselves are still working out how to handle these queries.
There is a further problem with the AI explanation as a catch-all. If AI tools were straightforwardly absorbing substance abuse treatment queries at scale, one would expect the impact to be roughly uniform — a consistent pull across keyword categories and volumes as people shift their searching habits. That is not what the data shows. Alcohol-specific terms fell an average of 54% while the largest general terms fell 18%. Mid-tier drug terms fell 45–56% while some recovery-framed terms actually grew. The declines are uneven in ways that do not map cleanly onto a general behavioral shift toward AI. Something more category-specific — and in the case of alcohol searches, more severe — is at work.
Mental Health Searches: An Interesting Parallel
Mental health-related search terms show a strikingly similar pattern of decline to those seen across substance abuse treatment searches. Across 430 mental health keywords tracked over the same period, more than four in five — 82% — declined year-over-year, with a volume-weighted average drop of nearly 30%. High-volume terms tell the story plainly: "therapy for depression" and "anxiety treatments," each averaging 110,000 monthly searches, are both down 18% year-over-year. "Bipolar disorder treatments" has fallen 63%. "Clinical depression" is down 70%. "Depression treatment" — 40,500 average monthly searches — down 45%.
This parallel matters for two reasons. First, it rules out the possibility that people seeking help for substance abuse have simply shifted toward mental health terminology — addressing root causes rather than searching for treatment directly. Mental health searches are falling just as sharply, so no such substitution effect is visible in the data. Second, and perhaps more significantly, it challenges the notion that declining SUD searches reflect declining need. Mental health need, by virtually every clinical and public health measure, has been moving in the opposite direction — the burden is growing, not shrinking. When searches collapse in parallel with a documented and ongoing rise in mental health burden, a different explanation is needed. The mental health data points to something disrupting search behavior itself — or the search behavior reporting — across both SUD and mental health — not a reduction in underlying need.
What This Year of Data Adds Up To
Stepping back from the individual numbers, a few things stand out from this dataset as a whole.
Alcohol treatment searches are significantly weaker than a year ago. The declines are not confined to a few obscure terms — they run across the core of how people search for alcohol help. "Alcoholic rehabilitation centers near me" is down 71%. "Alcoholism help near me" down 62%. "Alcohol programs near me" down 55%. And critically, this category did not participate in the partial recovery that showed up in broader treatment searches in the spring of 2026. Whatever that recovery reflected, it didn't reach alcohol-specific searches.
Drug treatment searches declined more moderately at the top, more steeply below. The highest-volume drug treatment terms fell roughly 18% year over year. But mid-tier drug terms fell significantly harder — 45 to 56% in several cases, representing losses of thousands of monthly searches per term.
Broader treatment and recovery searches declined less steeply — but still declined. The largest terms in this category, around 110,000 searches in April 2025, fell roughly 18% to around 90,500 — a loss of tens of thousands of monthly searches that should not be minimized by comparison to sharper drops elsewhere. Within this category, a minority of terms did show genuine growth: "Recovery center near me" grew 84%, "residential rehabilitation centers" grew 48%, and "detox centers" grew 22%. These are real signals, but they represent a small share of the category by volume, and they do not offset the overall downward direction. The partial overlap of these broad terms with non-addiction searches — physical rehab, medical treatment — likely provides some cushion in the numbers that purely addiction-specific terms don't have.
Luxury treatment searches collapsed across the board. Both luxury drug rehab and luxury rehab generally fell 70–80%, with the trend still declining heading into spring 2026. Florida-specific luxury searches were among the steepest drops in the dataset. Whether this reflects an economic shift, a reputational shift, or something else in the high-end treatment market isn't discernible from search data alone — but the signal is consistent and hard to dismiss.
In Summary
The larger question this data raises — but cannot answer — is what is driving these changes in search behavior. Fewer people searching for alcohol treatment doesn't necessarily mean fewer people need it. It might mean they're finding their way to help differently, through referrals, crisis lines, social channels, or AI tools. It might reflect broader social patterns. It might partly be a data artifact. All of those possibilities coexist, and this data alone cannot distinguish between them.
What complicates a single tidy explanation is the inconsistency of the declines across categories: alcohol searches fell far harder than drug searches; mid-tier terms collapsed while some broader terms held; certain recovery-framed language actually grew. That unevenness is not what a uniform Google reporting change or a clean AI migration shift would typically produce — it points to something more complex and nuanced, possibly a combination of factors operating differently across search types. Regardless, what Google's keyword trend tool shows fairly clearly is a broad decline in searches for substance abuse help — most dramatically in alcohol-related searches — that warrants further corroboration and explanation beyond what any single data source can provide.