Every store has its everyday dashboards – revenue, conversion, top products. But the catalog hides patterns no one built a view for. get_product_name_tokens lets you look exactly there: into the segments, dynamics and connections that don’t surface on their own – until you go looking.
Every product name in the store is a set of words. The tool breaks them into parts – each word is a separate token – and for each one it counts how much revenue the products containing it generate. Instead of looking at thousands of SKUs, you see which features of your catalog sell – and, more importantly, how fast they’re growing or declining.
tea herbal roses 100g
Gainers and losers
What’s rising, what’s falling – and why that matters more than revenue alone
Looking at tokens sorted by revenue shows you today’s state. To understand where the catalog is heading, you have to compare it with the previous period. The get_product_name_tokens tool with the compare_to=previous_period parameter returns the revenue change for each token – and that’s where it gets interesting.
In a specialty tea and infusions store I examined 90 days (February–May 2026) vs the previous 90 days (November 2025–February 2026). A few tokens moved so sharply they’re hard to ignore.
The biggest moves (sorted by absolute change, to filter out small tokens with a big %):
| Token | Revenue now | Change | Change % |
|---|---|---|---|
bundle | 124,304 EUR | +45,075 EUR | +56.9% |
pack | 106,010 EUR | −49,091 EUR | −31.7% |
starter | 105,334 EUR | −48,892 EUR | −31.7% |
detox | 105,146 EUR | +14,843 EUR | +16.4% |
mix | 67,197 EUR | −17,236 EUR | −20.4% |
botanical | 72,541 EUR | +13,775 EUR | +23.4% |
roses | 64,198 EUR | +12,344 EUR | +23.8% |
free | 79,469 EUR | −9,791 EUR | −11.0% |
caffeine | 76,730 EUR | −8,779 EUR | −10.3% |
adaptogens | 23,922 EUR | −5,023 EUR | −17.4% |
Structural signal: pack and starter both lose ~−49k EUR – the similar revenue and identical % change (−31.7%) suggest it’s the same “Starter Pack” product counted by two tokens at once. At the same time bundle grows +45k. Something is replacing the Starter Pack, but under a different name and probably in a different format.
Second signal: caffeine −10.3% (−8.8k EUR). The token is clearly losing – worth checking whether the “caffeine-free” product is selling worse, or simply whether the new SKUs don’t have that word in their name.
Yes – in DE starter −31.2%, set −31.4%, pack −16.9%. The same direction, a similar scale. Identical movement in two markets in the same window = a product signal, not a seasonal one. The starter pack as a format is losing traction regardless of market.
Rising
Falling
pack and starter often co-occur in the same product name (e.g. “Starter Pack”) – both refer to the same product – the same revenue decline counted by two tokens at once.Structural signal
The pack token loses −49k EUR abs. in PL – exactly like starter (−49k EUR, −31.7%). The identical scale of change confirms it’s the same “Starter Pack” product seen through two tokens. The same is happening in DE (starter −31.2%, set −31.4%). A simultaneous decline in both markets = not seasonality. Worth checking whether bundle (+57%) is replacing it, or whether these are entirely different customers.
An interesting detail
The matcha token grows +31.7% – faster than ginger (+13–14%). Matcha as an ingredient is outpacing ginger roughly 2×. It could be a signal of a new matcha line, or that customers associate it with higher quality.
Purchase intent
Purchase-to-view rate as a loyalty thermometer
Every token in the data has a ga4_purchase_to_view_rate field – how many purchases there were per product session containing products with that token. Values above 100% mean customers often buy repeatedly or return directly with intent to buy. This is a different kind of information than revenue alone – it’s a measure of the strength of purchase intent.
Sorted by PtV (from tokens with min. 20k EUR revenue):
| Token | PtV rate | Revenue | Profile |
|---|---|---|---|
multipack | 192% | 48,146 EUR | Returning, multipack |
refill | 182% | 94,175 EUR | Loyal, restocks supplies |
freeze-dried | 148% | 26,841 EUR | Premium, niche |
dried | 126% | 33,261 EUR | Known product |
leaves | 100% | 51,025 EUR | Regular customer |
Japanese | 115% | 43,557 EUR | Known product |
tea | 71% | 416,248 EUR | Core, mixed |
detox | 32% | 105,146 EUR | Exploring / comparing |
bundle | 21% | 124,304 EUR | Considering a purchase |
mix | 27% | 67,197 EUR | Exploring |
Interpretation: refill and multipack have PtV >100% – these are customers who don’t browse. They come in and buy. A product with the “refill” token doesn’t need conversion – it needs availability and an easy cart.
Meanwhile detox (PtV 32%) and bundle (21%) are products the customer hesitates over. Here the description, photos and reviews have a real impact on the buying decision.
The key to reading PtV
PtV >100% with rising revenue = a healthy segment of loyal customers. PtV >100% with falling revenue = a shrinking, ageing segment. Those who stayed – stay loyal. But the base isn’t renewing.
Value
Where the value really is – sorting by avg_price opens a different catalog
The default token sort is by revenue. Sorting by average selling price shows a completely different layer of the catalog – products customers buy little of but pay a lot for. And some of them grow much faster than their revenue share would suggest.
xxl: avg 239 EUR, 46 units, +22% vs the previous period. Token tub: avg 200 EUR, 65 units, +12%. Both sink in the revenue sort (11k and 13k EUR) – but they show there’s a premium segment ready to pay 10× more than for standard packaging.morning: avg 71 EUR, 214 units, +89% vs the previous period. These are morning sets – a higher basket than a single product, but the customer buys “the morning tea experience”. It’s nearly doubling. In the revenue sort it’s barely visible (15k EUR).infusion: avg 82 EUR, 22 units, +427%. The refined word “infusion” instead of plain “brew” → a premium segment. Compare: brew avg 14 EUR, 12,818 units. The same product category, completely different price positioning.allegro in PL: avg 88 EUR, 19 units, GA4=null, +354%. Products sold on Allegro sync to WooCommerce, but GA4 doesn’t see the transactions – the marketplace channel is growing while traffic analytics doesn’t know it exists.A hidden value pyramid
If you only look at tokens sorted by revenue, you see the mass – tea, rooibos, infusions. Sort by average price: a completely different catalog emerges – xxl (239 EUR), tub (200 EUR), morning sets (71 EUR). These aren’t niche products – they’re premium customers who are in your store, just invisible in the standard report.
Flavors and ingredients
Which flavors gain, which lose – and what it means for campaigns
Flavor and ingredient tokens are one of the few places where consumer trends are visible before they reach industry reports. Not through surveys – through what customers actually buy. In this store the data has a clear story to tell.
I’m grouping the flavor and ingredient tokens by direction of change:
Rising:roses +23.8% · fruity +23–38% · mint +22.6% · Ceylon +18% · white +16.5% · three-flavor (ginger) +26.5% · peach +3.8%
Falling:dried −12.9% · leaves −11.1% · flowers −7.6% · Japanese −6.7%
The pattern: Fruit and ginger are gaining – especially “indulgent” flavors (white, three-flavor, fruity, raspberry). Tokens associated with “basicness” are losing: tea leaves, dried ingredients, Japanese style. The customer is moving away from “healthy and simple” toward “healthy and enjoyable”.
In campaigns it’s worth boosting creatives with raspberry, fruity flavors and Ceylon tea. “Classic loose-leaf tea” creatives probably have a lower CTR than 6 months ago.
The flavor trend in brief
The customer is moving away from “healthy and simple” (loose-leaf tea, dried ingredients, Japanese style) toward “healthy and enjoyable” (ginger, raspberry, Ceylon, fruity flavors). The three-flavor token +26.5% is the clearest signal – specialty tea doesn’t have to be ascetic.
Summary
What you don’t see in a standard report
The same get_product_name_tokens call, four completely different questions – and each time a different finding that has no dashboard of its own. The direction of change reveals that the Starter Pack is structurally losing in both markets. PtV says where the customer buys without browsing. The price sort uncovers a hidden premium segment. The flavors show a shift in preference: from “healthy and simple” to “healthy and enjoyable”. None of these patterns pops out of a top-products report on its own – because no one knew in advance it was worth looking for.
The main takeaway
The catalog has patterns no one planned for. Tokens find them.
Most e-commerce analyses answer questions someone formulated in advance – and for which someone built a view in advance. Tokens work differently: instead of confirming what you already know, they reveal what’s there but you weren’t looking for. Unexpected momentum, a hidden premium segment, a shift in flavor preferences – these are things that have been in the data for months but have no dashboard of their own.
The biggest finding from this analysis: the Starter Pack loses ~49k EUR with a PtV above 100% – loyal customers stay, but the segment isn’t renewing. This is happening in both markets at once. It’s a question of product and acquisition, not seasonality. And it only appears when you ask about the direction of change – not from a top-products report.
DataOrganizer · MCP
Connect DataOrganizer and ask about token gainers and losers. The comparison with the previous period takes a second – the conclusions last for weeks.