StrategyAugust 16, 2026·5 min read

Penetration pricing and KVIs: where to go aggressive

Penetration across the whole catalog is a margin cut. The winners go deep on 5-10% of SKUs - the KVIs - and hold the line elsewhere. How to find your real KVIs and how aggressive to be.

Penetration pricing across an entire catalog is not a strategy - it is a margin cut with a business case attached. The retailers who make aggressive pricing pay for itself do one thing differently: they go deep on a narrow, deliberately chosen set of SKUs and hold the line everywhere else. Retailgrid exists to make that split operational, because the hard part is not deciding to be aggressive. It is deciding exactly where.

Why KVIs are the right target

Key value items are the products shoppers use to judge whether your store is expensive. A small set - often 5-10% of SKUs - drives the majority of price perception. Nobody remembers what you charged for a niche accessory. Everyone remembers the price of the item they check every visit.

That asymmetry is the whole opportunity. Aggressive pricing on KVIs buys price perception at a fraction of the margin cost of a broad discount, because the remaining 90% of the catalog keeps earning.

Identifying your real KVIs

Most retailers think they know their KVIs. Most are working from a list assembled years ago by intuition. Rebuild it from data:

  • Search and comparison frequency. Which SKUs get checked before purchase, on-site and off?
  • Purchase frequency and basket presence. Repeat-purchase staples anchor perception more than occasional buys.
  • Price elasticity. Measured demand response, not assumed sensitivity.
  • Competitor coverage. A product only functions as a KVI if competitors stock it and shoppers can compare.
  • Basket pull. Does the item bring a trip that includes profitable products?

Product role classification turns this into a maintained segmentation - traffic drivers, margin builders, and tail items - so pricing rules apply by role rather than by category habit.

How aggressive is aggressive enough

On KVIs, shallow discounting is wasted money. A 3% gap below the competitive band is invisible to shoppers and still costs you margin. Either position clearly - typically 10-20% below the band midpoint, or at competitor minimum - or do not bother.

This requires knowing where the band actually sits today. Competitor price tracking refreshed on a four-hour cycle tells you the live position; a quarterly benchmarking study tells you where the market was last season. Competitive pricing rules let you define which competitors count as reference points, so you match the retailers your customers actually compare you to rather than every listing on a marketplace.

Protecting the other 90%

Aggression on KVIs only works if margin recovery elsewhere is deliberate. Set the tail catalog on markup rules with a firm floor, price margin builders on optimization rather than competitor matching, and cap how far competitor-triggered moves can spread beyond the KVI list.

The failure mode is contamination: a rule written loosely lets KVI logic bleed into adjacent SKUs, and within a quarter the whole category is priced at competitor minimum. Scope every aggressive rule to an explicit SKU set with an expiry date.

Keeping position without manual work

KVI pricing decays fast. Competitors reprice, your position drifts, and price perception erodes without anyone noticing. Dynamic pricing maintains the intended position automatically within your caps and floors - reacting to competitor moves, stock levels, and demand rather than waiting for the next review meeting.

Measure perception, not just units

Track price index against your reference set on the KVI list specifically, basket size and margin mix on the trips those KVIs generate, and realised category margin overall. If KVI index improves while category margin holds, the strategy is working. If both fall together, the aggression has leaked beyond the list.

Ready to improve your pricing strategy? Book a demo with Retailgrid.

See the agentic pricing platform behind the writing.

A 20-minute walkthrough of Retailgrid on a real retail dataset. No signup. No sales script.