Contact Us
Contact Us
Iris
Give me a short assortment story I can bring to this week's review: hindsight, clustering confidence, option health, and where to focus.
Iris · assortment snapshot, this season

Assortment snapshot · this season

  • Hindsight: 5 carry-forwards confirmed, 3 styles flagged as misses not to repeat
  • Clustering: Current store clustering confirmed at 91% confidence
  • Option health: 3 departments show gaps vs. strategy, concentrated in Denim and Bottoms
  • Buy: Confidence in the current buy sequence sits just below threshold, manual review recommended
  • Attention: two departments need focus this week

Three asks for merchandisers (end of day):

  • Review the flagged Denim and Bottoms option-count gaps
  • Confirm the buy plan drops below the confidence threshold
  • Check that carry-forward decisions are reflected in the current line plan
AssortSmart®

Decide What to Carry, How Much, and Where with Iris

Only With Impact Analytics
Iris, the AI assortment planning agent inside AssortSmart, helps merchandisers decide what to carry, how much to buy, and where each product should go. Ask a question in plain language, and Iris investigates demand, store performance, clustering, option counts, and buy plans, then explains the evidence and stages the recommended change for approval.
Contact Us
Contact Us
Trusted by Market-Leading Global Brands
THE AGENT AT WORK

Every Assortment Question, Answered with Iris

Ask the AI assortment planning agent about any category, product, store group, or buy decision. Iris investigates the plan, explains what is driving the result, and stages the next action inside AssortSmart.

Option Gaps

Which Categories Need More—or Fewer—Options?

Iris checks the developing range against assortment strategy, ranks the largest gaps and overages, and shows your team where the line needs attention before approval.

‍

  • Identifies categories above or below planned option counts.
  • Quantifies the gap by product role and department.
  • Returns a prioritized worklist for review inside AssortSmart.
Iris
Which departments have option-count gaps against strategy this week?
Iris · option-count health, this week

This week · option-count health

  • 3 departments flagged
  • Gaps concentrated in Womenswear Bottoms and Denim

The largest gap sits in departments where Hero-role SKUs are under-represented against the strategy target. I'd prioritize those first. Want the ranked list?

Show the list, ranked by gap size.
Iris · ranked by gap size
DepartmentIssuePriority
Denim4 options short of Hero-role targetHigh
W. Bottoms · TrousersDepth below recommended targetHigh
W. Tops · KnitsFiller role over-representedMedium

Suggested next steps: generate placeholders for the high-priority gaps first, then rebalance Filler in Knits.

I won't change the line plan without your review. Commit or discard each line when you're ready.

Sell-Through Analysis

Understand Why a Category Sold Below Plan

When sell-through misses plan, Iris separates demand, depth placement, store clustering, product roles, and newness to identify what actually drove the result.

‍

  • Separates demand problems from allocation or depth-placement issues.
  • Identifies the store groups and product roles driving the miss.
  • Recommends the next action inside AssortSmart.
Iris
Why is sell-through soft in the outerwear even though we increased depth this season?
Iris · AssortSmart

Outerwear · sell-through vs. strategy

  • Depth: Depth increase was concentrated in lower-velocity clusters
  • Mix: A larger share of the added depth landed in Filler roles rather than Hero
  • Fabric: Newness skewed toward fabrics with softer early sell-through signals last season

Conclusion: This looks like a depth-placement issue. More units, but concentrated in the wrong clusters and roles, not a demand problem.

Next in AssortSmart

Review the cluster-level depth split for Outerwear and rebalance toward your higher-velocity clusters.

Range Scenarios

Build the Range Around Your Business Target

Give Iris a margin, sell-through, or inventory target, and the AI assortment planning agent builds a recommended range and depth plan around it while preserving locked decisions and business constraints.

‍

  • Optimizes breadth and depth against the target you set.
  • Preserves locked carry-forwards and confirmed decisions.
  • Stages the complete recommendation before anything is committed.
Iris
If I want to protect margin here while still growing sell-through, what would the depth split need to look like?
Iris · AssortSmart

I built a depth-and-option-count distribution across Denim's clusters that targets your margin floor while growing sell-through, honoring your locked carry-forwards. Here's the staged plan.

Review it before anything commits.

If you'd rather weight this toward sell-through over margin, tell me and I'll rebuild it with that priority.

Store Clustering

Know How Much Confidence to Place in Every Store Cluster

Iris evaluates the statistical strength of each store cluster, explains the demand and product patterns behind it, and flags stores that may distort the grouping before assortment decisions are built on top of it.

‍

  • Shows the confidence level behind each store grouping.
  • Explains the demand and product patterns driving each cluster.
  • Flags stores that may distort the grouping before the range is built.
Iris
How confident are we in the current clustering, and what's driving the High Street cluster?
Iris · AssortSmart

Store clustering · 91% confidence

  • Lens: store performance and product affinity
  • High Street cluster: mid-tier AUR spread, strong stability score, largest store count
  • Two stores flagged as outliers: limited trading history, may distort the grouping

I'd recommend reviewing the outliers before you lock this in. Want the outlier list, or a breakdown by another cluster?

Which Categories Need More—or Fewer—Options?

Iris checks the developing range against assortment strategy, ranks the largest gaps and overages, and shows your team where the line needs attention before approval.

‍

  • Identifies categories above or below planned option counts.
  • Quantifies the gap by product role and department.
  • Returns a prioritized worklist for review inside AssortSmart.
Iris
Which departments have option-count gaps against strategy this week?
Iris · option-count health, this week

This week · option-count health

  • 3 departments flagged
  • Gaps concentrated in Womenswear Bottoms and Denim

The largest gap sits in departments where Hero-role SKUs are under-represented against the strategy target. I'd prioritize those first. Want the ranked list?

Show the list, ranked by gap size.
Iris · ranked by gap size
DepartmentIssuePriority
Denim4 options short of Hero-role targetHigh
W. Bottoms · TrousersDepth below recommended targetHigh
W. Tops · KnitsFiller role over-representedMedium

Suggested next steps: generate placeholders for the high-priority gaps first, then rebalance Filler in Knits.

I won't change the line plan without your review. Commit or discard each line when you're ready.

Understand Why a Category Sold Below Plan

When sell-through misses plan, Iris separates demand, depth placement, store clustering, product roles, and newness to identify what actually drove the result.

‍

  • Separates demand problems from allocation or depth-placement issues.
  • Identifies the store groups and product roles driving the miss.
  • Recommends the next action inside AssortSmart.
Iris
Why is sell-through soft in the outerwear even though we increased depth this season?
Iris · AssortSmart

Outerwear · sell-through vs. strategy

  • Depth: Depth increase was concentrated in lower-velocity clusters
  • Mix: A larger share of the added depth landed in Filler roles rather than Hero
  • Fabric: Newness skewed toward fabrics with softer early sell-through signals last season

Conclusion: This looks like a depth-placement issue. More units, but concentrated in the wrong clusters and roles, not a demand problem.

Next in AssortSmart

Review the cluster-level depth split for Outerwear and rebalance toward your higher-velocity clusters.

Build the Range Around Your Business Target

Give Iris a margin, sell-through, or inventory target, and the AI assortment planning agent builds a recommended range and depth plan around it while preserving locked decisions and business constraints.

‍

  • Optimizes breadth and depth against the target you set.
  • Preserves locked carry-forwards and confirmed decisions.
  • Stages the complete recommendation before anything is committed.
Iris
If I want to protect margin here while still growing sell-through, what would the depth split need to look like?
Iris · AssortSmart

I built a depth-and-option-count distribution across Denim's clusters that targets your margin floor while growing sell-through, honoring your locked carry-forwards. Here's the staged plan.

Review it before anything commits.

If you'd rather weight this toward sell-through over margin, tell me and I'll rebuild it with that priority.

Know How Much Confidence to Place in Every Store Cluster

Iris evaluates the statistical strength of each store cluster, explains the demand and product patterns behind it, and flags stores that may distort the grouping before assortment decisions are built on top of it.

‍

  • Shows the confidence level behind each store grouping.
  • Explains the demand and product patterns driving each cluster.
  • Flags stores that may distort the grouping before the range is built.
Iris
How confident are we in the current clustering, and what's driving the High Street cluster?
Iris · AssortSmart

Store clustering · 91% confidence

  • Lens: store performance and product affinity
  • High Street cluster: mid-tier AUR spread, strong stability score, largest store count
  • Two stores flagged as outliers: limited trading history, may distort the grouping

I'd recommend reviewing the outliers before you lock this in. Want the outlier list, or a breakdown by another cluster?

FEATURES

Every Assortment Decision Explained Before You Approve It

The AI assortment planning agent reads the full assortment lifecycle, from last-season performance and store clustering through option counts and the buy. Every recommendation is grounded in evidence, every exception is visible, and every change stays within the guardrails your team controls.

Last Season Informs the Next Assortment

Every product is evaluated against prior sales, margin, sell-through, and store-group performance. Iris recommends what to carry forward and clearly identifies what should not return.

Stores Grouped by How They Actually Sell

Stores are clustered by demand and product performance, with a confidence score behind each grouping so teams understand how reliable it is before building the range.

Option Gaps Surfaced Before Approval

Iris checks option counts, product roles, and range depth against assortment strategy, flagging gaps and overages before the line is committed.

Every Answer Traceable to Its Source

Every answer identifies the planning stage, business level, and data behind it. Missing information is stated explicitly rather than filled with assumptions.

Draft, Review, Approve Every Change

Every recommended change arrives as a draft your team reviews before the plan updates. Routine, low-risk changes can run automatically within the limits your administrators set.
Get Started

From First Assortment Question to an Approved Product Range

The AI assortment planning agent follows the same transparent process every time: investigate the assortment, explain the evidence, recommend the next move, and stage the change for approval.
01

Investigate

Iris reads prior-season performance, store clustering, option counts, range strategy, and buy plans to identify where the assortment is drifting from demand or strategy.
02

Recommend

AssortSmart's planning engines build the recommendation, with the expected outcome, affected products, and supporting evidence attached.
03

Approve

Every change stays staged until your team approves it. Routine, low-risk decisions can execute automatically only within administrator-defined guardrails.
01

Investigate

Iris reads prior-season performance, store clustering, option counts, range strategy, and buy plans to identify where the assortment is drifting from demand or strategy.
02

Recommend

AssortSmart's planning engines build the recommendation, with the expected outcome, affected products, and supporting evidence attached.
03

Approve

Every change stays staged until your team approves it. Routine, low-risk decisions can execute automatically only within administrator-defined guardrails.

Frequently Asked Questions

What is an AI assortment planning agent?

An AI assortment planning agent helps merchandisers decide which products to carry, how much to buy, and where those products should go. Iris works inside AssortSmart, using demand, store performance, clustering, option counts, and buy plans to explain and recommend assortment decisions in plain language.

How does the AI assortment planning agent work?

Iris evaluates prior-season performance, store clustering, assortment strategy, option counts, and buy plans. It identifies gaps, explains what caused them, and stages recommended changes inside AssortSmart for merchandiser review.

Does the assortment planning agent build my assortment for me?

Yes, it can build the draft. Give Iris a margin, sell-through, or inventory target, and it can recommend range breadth, depth, and store-group distribution while respecting locked carry-forwards and business rules. Your team reviews the plan before it is committed.

How is an AI assortment planning agent different from assortment planning software?

Traditional assortment planning software holds the plan and waits for users to analyze it. An AI assortment planning agent actively investigates the plan, identifies gaps, explains why performance changed, and drafts the recommended action for review.

Autonomous AI Agent Orchestration

Only With Impact Analytics
Rethink how your business makes decisions. Replace disconnected AI with agentic intelligence that moves every function toward the same business outcome. Discover how Impact Analytics helps you lead the shift to a fully orchestrated, autonomous enterprise.
Contact Us
Contact Us