Personal trend research · Public-domain forecasting · AI imagery to match
YakLab scans live web signals across 55 fashion categories — runway, retail, social, palette direction — and reads the shops themselves, from your own photographs. Builds research-backed briefs. Decomposes them into themed stories with AI plates, textile prints, and CAD shape matches. Export as an editorial trend book PDF.
Inside the app: YakTREND research, shop reports, briefs, stories, plates, prints — one sidebar. No sprawl, no setup.
55 personal trend scans across womenswear, menswear, kids, swim, denim, knitwear, lingerie, home, licensing and more. Each returns a 20-page snapshot from live web sources — runway, retail, social signal, palette direction, fading trends. Yours forever; re-read whenever.
From the photographs and inspiration shots you take on a trip, through to a complete and organised trend analysis. Every image is read on its own against a fixed vocabulary, so trends form by measured similarity rather than impression, and every ranking is a number you can challenge. Then take any trend through to a brief — mood imagery, textile prints and graphics, built from what those shops are actually selling.
Anchor work in research. Write your own concept or build straight from a YakTREND scan. Google-grounded research, reference moodboards, named palette and material direction — all in one editable record.
Each brief decomposes into themed stories. Each story gets a name, tagline, 60–90 word summary, and an 8-colour named palette. The narrative backbone of the trend book.
Six visual plates per story — atmosphere, material, detail, hero, worn, interior. Tuned to the story's palette, audience and category. Standard or Refined — your call per plate.
Six prints per story — botanical, geometric, stripe, painterly, placement, texture. Category-aware: tailoring pulls micro-checks; swim pulls tropical botanicals; kids pulls conversational; home pulls cushion-scale. Production-ready surface design.
One click → editorial trend book. Cover plus up to four story spreads (palette, plates, prints). Shareable, printable, never re-renders. Two templates — Luminous and Editorial — so the same research can be presented two ways.
5,400 vector flats across womenswear, menswear, kids and home, searchable by category, subcategory and market. Every fashion-category report surfaces 8–12 directional shapes matched to its signals. Apply a story's palette for a colourway sheet, or unlock a shape as PNG, as a front-and-back set, or as layered .ai + .svg to design onto.
Five screens, one workflow: from a public-domain trend signal — or a day’s shopping report — to a finished trend book.
Pick a research area, set a time horizon. YakTREND scans live web sources — runway, retail, social, palette direction — and returns a research-backed report anchored to today's date. Free to re-read forever.
Shoot a folder per retailer as you go — which shops carry a trend is part of what decides its rank. Every image is read on its own against a fixed vocabulary, then grouped by measured similarity, and you choose how many trends to pull out, up to eight. Take any one through to a brief and the palette is measured off the garments, not invented: the prints that come back are built from what those shops are actually selling.
Build a brief from your YakTREND scan, from a Shop Report trend, or write your own concept. Gemini-grounded research locks in tone, palette and materials. Then decompose into 3–5 themed stories — each with name, tagline, summary, and an 8-colour named palette.
Each story gets six visual plates (atmosphere, material, detail, hero, worn, interior) and six textile prints (botanical, geometric, stripe, painterly, placement, texture). Category-aware: tailoring pulls micro-checks; swim pulls tropical botanicals; kids pulls conversational.
Export an editorial-format trend book: cover plus up to four story spreads (palette, plates, prints). For fashion-category scans, YakLab also surfaces 8–12 directional vector CADs from a 5,400-shape library — matched to the trend's signals and ready to design onto.
The tool sits in the gap between a trend agency and an AI image tool — research-backed, category-aware, and pay-per-use.