Ones to watch
Perth cafe trends with the strongest early signals, scored out of 100.
The early signal score
Perth's scores come from Google reviews, cafe menus and searches. We don't read Perth cafes' posts yet, so the parts built on posts (creators and likes) have no reading here.
A number from 0 to 100 for how strong a trend's early signals are today. It is new and unproven. It is not a prediction and not a chance of anything: we do not have enough history yet to test one.
- Spread 22% of the score
- How many unrelated cafes posted it, and across how many suburbs.
- Momentum 22% of the score
- Whether more of the same cafes post it now than before, or more cafes and creators than its own history predicts.
- Search 16% of the score
- How many people in Perth search for it, and whether that is growing.
- Creators 10% of the score
- How many independent food creators posted it.
- Confirmation 10% of the score
- How many different kinds of evidence agree: cafe posts, creators, menus, reviews, press, searches.
- Engagement 8% of the score
- Whether posts about it get more likes than the cafe's usual post.
- Names 7% of the score
- Whether cafes are giving it their own names.
- Freshness 5% of the score
- Whether we first saw it recently and it is still being posted.
A part with too little behind it is left out and marked "not enough yet", so missing data never counts against a trend. Anything seen at fewer than 3 cafes is held at 25 or under, and anything no cafe has posted yet at 10.
Every day we record each score and the figures behind it, for the trends that go nowhere too, so the score can be checked against what happens and corrected.
Version 0.1, set on 12 Nov 2026. How the score works, in full
The radar
43 trends to watchReading the radar
Each drawing is a trend. Across: how many Perth cafes posted it in the last 90 days (each step is about twice as many). Up: whether more of the same cafes have it now than in the 90 days before.
Up and to the left is the corner to watch: growing, on few cafes so far.
Past the dashed line are trends already on more than 1 in 10 of the 632 Perth cafes we read. They are not ones to watch any more, so they are listed apart, not ranked.
Reading the radar
Each drawing is a trend. Across: how many Perth cafes posted it in the last 90 days (each step is about twice as many). Up: whether more of the same cafes have it now than in the 90 days before.
Up and to the left is the corner to watch: growing, on few cafes so far.
Past the dashed line are trends already on more than 1 in 10 of the 632 Perth cafes we read. They are not ones to watch any more, so they are listed apart, not ranked.
Pick a drawing to see the trend. Figures to 14 Nov 2026.
Right: more cafes have it. Up: more of the same cafes than in the 90 days before.
bigSlowingMore cafes Growing
- 1Mont Blanc coffee
- 2Cold foam
- 3Hot honey
Mont Blanc coffee
31 cafes · 1 of the same cafe, down from 2
Ranked by early signalReading the list
Each row is a trend in its Emerging or Rising stage, with its strongest reason and its early signal score out of 100.
“Up from” compares the same cafes over the last 90 days with the 90 days before, so reading more cafes never looks like a trend taking off.
Micro means fewer than 10 cafes have posted it in the last 90 days in every city we track. The tag goes by itself once 10 or more cafes in any city have it.
Trends already on more than 1 in 10 Perth cafes are named under the list, not ranked.
Reading the list
Each row is a trend in its Emerging or Rising stage, with its strongest reason and its early signal score out of 100.
“Up from” compares the same cafes over the last 90 days with the 90 days before, so reading more cafes never looks like a trend taking off.
Micro means fewer than 10 cafes have posted it in the last 90 days in every city we track. The tag goes by itself once 10 or more cafes in any city have it.
Trends already on more than 1 in 10 Perth cafes are named under the list, not ranked.
“Up from” compares the same cafes with the 90 days before.
43 trends shown
Show all 43 trendsShow fewer trends
Already big, so not ranked
Run a cafe? See what Perth's frothing for
Questions people ask
What new cafe trends are emerging in Perth?
The Froth has 43 trends on its Emerging board as of 14 Nov 2026: trends in their Emerging or Rising stage, 9 of them small enough to be tagged Micro. The strongest early signals are Mont Blanc coffee (66), Cold foam (65) and Hot honey (58), on our early signal score out of 100.
How does the early signal score work?
It is a number from 0 to 100 built from eight parts: how many unrelated cafes post a trend, whether more of the same cafes post it than before, Perth searches, food creators, how many kinds of evidence agree, likes, the names cafes give it, and how recently it was first seen. A part with too little behind it is left out, and anything seen at fewer than 3 cafes is held at 25 or under.
Is the early signal score a prediction?
No. The score (version 0.1) is new and unproven. It sums up today's signals and shows its reasons; it is not a forecast and not a probability. We do not yet have enough history to test one, so we record every score daily, including for trends that go nowhere, to check it against what happens.
What does the Micro tag mean?
Micro means fewer than 10 cafes have posted it in the last 90 days in every city we track. The tag goes by itself once 10 or more cafes in any city have it. It is the same trend either way: every name it goes by, a share of cafes, a stage and a list of places to get it.
Facts to quote
- 43 trends are on The Froth's Perth Emerging board as of 14 Nov 2026, each in its Emerging or Rising stage; 9 are tagged Micro (fewer than 10 cafes in the last 90 days in every city).
- Mont Blanc coffee has the strongest early signals, with a score of 66 out of 100 (31 cafes in the last 90 days).
- Of the Micro trends, Croffle scores highest at 31 out of 100.
- Matcha (1 in 2 cafes), Iced latte (1 in 3 cafes), Cold brew (1 in 7 cafes), Strawberry matcha (1 in 9 cafes) are already on more than 1 in 10 Perth cafes, so they are listed apart rather than ranked.
- The early signal score (version 0.1, set 12 Nov 2026) is built from eight parts and is not a prediction: it is new and has not been tested against outcomes yet.
- A trend seen at fewer than 3 Perth cafes is held at 25 or under, whatever its searches say.
Updated
How the score works
A number from 0 to 100 for how strong a trend's early signals are today. It is new and unproven. It is not a prediction and not a chance of anything: we do not have enough history yet to test one.
- Spread 22% of the score
- How many unrelated cafes posted it, and across how many suburbs.
- Momentum 22% of the score
- Whether more of the same cafes post it now than before, or more cafes and creators than its own history predicts.
- Search 16% of the score
- How many people in Perth search for it, and whether that is growing.
- Creators 10% of the score
- How many independent food creators posted it.
- Confirmation 10% of the score
- How many different kinds of evidence agree: cafe posts, creators, menus, reviews, press, searches.
- Engagement 8% of the score
- Whether posts about it get more likes than the cafe's usual post.
- Names 7% of the score
- Whether cafes are giving it their own names.
- Freshness 5% of the score
- Whether we first saw it recently and it is still being posted.
A part with too little behind it is left out and marked "not enough yet", so missing data never counts against a trend. Anything seen at fewer than 3 cafes is held at 25 or under, and anything no cafe has posted yet at 10.
Growth is only ever measured on the same cafes in both periods, so reading more cafes never looks like a trend taking off. A breakout in searches is scaled by the searches behind it: a jump from almost nothing adds almost nothing.
The tag beside a score says how much it rests on: good evidence is six or more parts with a reading and at least 10 cafes; some evidence is four parts and 3 cafes; anything less is thin.
Every day we record each score and the figures behind it, for the trends that go nowhere too, so the score can be checked against what happens and corrected.
Version 0.1, set on 12 Nov 2026.