Category Ranking

Category Optimization

Let the system order your category pages by real sales performance.

Instead of manually deciding which product sits at the top of a category page, the system re-ranks every hour by weighing sales velocity, stock levels, and seasonality together. Sold-out items sink, true sellers rise. And if a result isn't what you expected, one click rolls it back.

Smart Scoring Engine

A 0-100 performance score for every product. Sales velocity, last sale date, stock status, and price signals are weighed together — with fair, reliable results even in small categories.

  • Sales velocity — last week and last month combined
  • Days since the last sale
  • Score drops when a critical size sells out
  • Guards against inflated scores on thin data
  • Balance between in-category and brand-wide performance
Product Score Detail
Crepe Shawl Collar Dress · M
Total Score
71
Sales Velocity
weight 35%84
Last Sale Date
weight 20%92
Stock Status
weight 25%65
Discount Signal
weight 20%42
Fair scoring: In categories with few products, artificially high scores are prevented — balanced against the brand average (actual 76 → adjusted 71).

Apply to Store (Manual Approval)

Scores are computed every hour, but nothing is written to your store without your approval. Update the whole category or selected products in one click — and roll back just as easily.

  • Works with Tsoft, Ticimax, Shopify, and ikas
  • Pin to top/middle/bottom — manual spotlighting
  • Category boundary protection — adjacent categories stay intact
  • Scheduled applies — run automatically at the hour you choose
  • Preview + change comparison table
Apply — Preview
Women · Dresses (Shopify)
ProductOldNewChange
Crepe Shawl Collar Dress#8#2-6
V-Neck Silk Shirt#14#4-10
Wide-Leg Trousers#3#7+4
Pleated Mini Skirt#22#12-10
Tie-Front Blouse#5#18+13
A backup is taken before applying — you can revert with one click from the Outcome Measurement page.

Outcome Measurement — Did We Make the Right Call?

24-72 hours after every apply, the sales impact is measured. Positive results carry on; negative ones can trigger an automatic rollback (optional).

  • Before/after sales comparison for every apply
  • A/B test mode — split the category and compare
  • Negative result → automatic rollback (with your approval)
  • Weekly results summary in Slack
Outcome Measurement
Apply #4218 · 48 hours later
▲ Positive
Hourly Units Sold+18.4% vs before
Apply
72h agoApply (48h ago)Now
Before/hr
3.2
After/hr
3.8
Decision
Continue

PLP Comparison — Before/After View

Compare two rankings side by side: the current PLP versus the proposed one. See which product moves where — and what it's worth — before you apply.

  • Side-by-side list — current order vs. new order
  • Position change arrows (▲ +5 / ▼ -3)
  • Pull-back suggestions for out-of-stock products
  • Estimated sales impact (from historical analogues)
PLP Comparison
Women · Dresses
5 changes
Current Rank
Suggested Rank
#1Crepe Shawl Collar Dress
=
#1
#2V-Neck Silk Shirt (out of stock)
#12-10
#18Tie-Front Blouse
#2+16
#6Wide-Leg Trousers
#3+3
#4Pleated Mini Skirt
#8-4
Estimated impact: +6-9% category revenue (historical average of similar applies).
Highlights

The module at a glance

Automatic PLP ranking — driven by product performance
Sold-out products drop to the bottom automatically
Sales impact measured after every apply
Automatic rollback on negative results (optional)
Use Cases

Where does it help in practice?

Case 1

The summer collection just dropped — the rug category needs a reshuffle

Outcome

Summer products score higher → they rise on the PLP automatically. Winter stock stays down.

Case 2

Out-of-stock products are wasting traffic at the top of the category

Outcome

Sold-out items sink automatically — and climb back when stock returns.

Case 3

Manually spotlighting a new product launch

Outcome

Pin top — a guaranteed manual top slot; pin bottom — manually kept in the back.

Demo

Try the Category Ranking module with your own brand

In a 30-minute demo session, we'll show you how the module works on your own data.