Tracking Demand for Aesthetic and Pharmaceutical Beauty Products With Google Trends
Google Trends is free, requires no API key, and most people use it wrong. I've spent the better part of three years pulling trend data for aesthetic dermatology and cosmetic pharmacology brands, and the difference between useful output and noise mostly comes down to how you structure your queries and what you do with the results afterward. The basic workflow is straightforward. You go to trends.google.com, switch the setting to "Web Search" or "Image Search," pick a region and date range, and enter keywords. Where people trip up is in step three. Picking the right keywords determines whether you get clean data or a garbage plot that looks like a heart attack on an EKG. Here's how I actually run these queries.
I start by building a master list of terms. For aesthetic pharmacology, that means breaking it into subcategories: injectables (botulinum toxin, dermal fillers, bioestimulators like PLLA and PCL), topical pharmacology (retinoids, tranexamic acid, glutathione, peptide complexes), oral supplements with cosmetic indication (collagen peptides, silimarina, astaxanthin), and procedure-linked terms (laser hair removal aftercare, PRP recovery, thread lift downtime). Each subcategory gets its own trend query rather than lumping everything into one search, which distorts the relative interest scores. From there, I pull data in six-month windows going back five years. Google Trends gives you a normalized interest score from zero to one hundred, where one hundred is the peak search volume for that term in the selected region and timeframe. The raw numbers mean nothing on their own. What matters is the ratio between terms and the seasonality patterns that emerge when you overlay them. I export the data as CSV and run it through a simple Python script that calculates year-over-year growth rates and flags anomalous spikes. A spike during the first week of January is almost certainly New Year resolution noise. A spike in late September across multiple related terms usually signals prep-for-wedding-season demand or summer recovery cycles in the northern hemisphere. The script also flags when a term drops below three percent relative interest for two consecutive quarters, which for most commercial applications means the market has either saturated or migrated to a different search behavior.
One thing that catches people off guard: Google Trends relative interest is not search volume. A term at 80 in a small market like New Zealand can represent fewer total searches than the same term at 30 in Brazil. If you're making business decisions about inventory or ad spend, you need to cross-reference with actual search volume data from Keyword Planner or SEMrush. I learned this the hard way in 2023 when a client was convinced that demand for exosome therapy was skyrocketing based on Trends data alone. The relative score had climbed to 65, but absolute monthly searches in their target market were under two thousand. The trend looked impressive until you looked at the denominator. Here's the edge case I wish I'd known about earlier. Google Trends automatically groups semantically similar queries under a single term unless you tell it not to. For aesthetic pharmacology, this causes a specific problem: terms like "Botox," "botox injection," "botox treatment," and even "botulinum toxin" get collapsed into one data series. If you're tracking competitive positioning between brand names and generic pharmacological terms, the default grouping hides the distinction. I solved this by using the "Related queries" breakdown feature and manually separating branded from unbranded interest. The workaround is tedious but necessary. You pull the related queries report for each term, identify which variants carry meaningful commercial signal, and then run those as individual queries with the grouping disabled. It adds roughly forty-five minutes to what would otherwise be a twenty-minute research session, but the data you get back is actually usable instead of misleading.
Get the Full Details
Another counter-intuitive detail that matters: the geo-filter on Google Trends has a hard cutoff at the second level for most countries. In the United States, you can drill down to metropolitan statistical areas. In Canada, you're stuck at the province level. In Germany, you get states but not cities. If your aesthetic pharmacology product targets a specific urban demographic, the resolution might not be granular enough. I've had cases where a term showed flat nationwide interest but the "Where to watch" equivalent for search data revealed that ninety percent of the volume came from two metro areas. The workaround is to query each major metro area individually by setting the location to "Greater London" or "New York City" type regions that Google recognizes. This multiplies your query count but gives you the geographic specificity that national-level data obscures. Limitations you should keep in mind. Google Trends has a maximum lookback of five years for most categories, though some higher-volume terms go back further depending on Google's internal data retention. More importantly, the tool does not distinguish between commercial intent and informational curiosity. Someone searching "how does hyaluronic acid filler work" gets the same weight as someone searching "buy hyaluronic acid filler near me." If you're using this for market sizing or revenue forecasting, you'll overestimate demand unless you layer in intent-segmented data from a proper keyword research platform. Google Trends is best used for understanding timing, seasonality, and comparative interest—not for determining how many units you'll move next quarter.
There's also the category filter problem. If you set the category to "Health" when pulling aesthetic pharmacology data, Google narrows the index to health-related pages only. That excludes e-commerce, forums, and social media discussions where a lot of consumer aesthetic decision-making actually happens. Leaving the category as "All categories" gives you broader but messier data. I usually run both and compare the results. The difference tells you how much of the search behavior is clinical versus consumer-driven, which is useful information in itself. For anyone actually doing this work regularly, I'd recommend setting up a recurring weekly export routine rather than pulling data ad hoc. The manual process of adjusting date ranges, checking for grouping artifacts, and exporting clean CSVs takes about twenty minutes per term per week. Over a twelve-month period, that compounds into something significant. A basic script using the pytrends library in Python handles the automation, but you need to rotate user-agent strings and throttle requests to avoid temporary blocks. Google doesn't announce these limits publicly, but after about fifty queries in a single day from the same IP, the tool starts returning partial or errored responses. Space your pulls out across multiple days if you're running a large keyword list. The data itself is free. The insights require discipline. Most people who try aesthetic pharmacology trend analysis quit after the first month because the output looks vague. It's vague until you constrain the query parameters, account for grouping artifacts, and cross-reference with absolute volume data. Do that and the tool becomes genuinely useful for deciding when to launch a campaign, which subcategory to prioritize, and whether a term's growth is structural or seasonal.
