Retail Pricing
Price optimization software
A buyer’s guide to the retail pricing system: the three modules the category sells, the difference between a rule engine and a demand model, seven things to evaluate, and one worked example where the same pricing rule makes money on one item and loses it on another.
Two different purchases share this vocabulary
This page is about the system a retailer buys to set the prices in its own stores: the RELEX and Revionics class of product, sitting between the item file and the POS, producing a price file that goes to the register. It is an operational system purchase with an integration project attached.
A supplier selling into retail is buying something else entirely. A brand does not set the shelf price, so it is not buying a price file. It is buying measurement: base versus promoted elasticity, promotion return after deductions, and price index against a competitive set. That is a different product and it has its own page, pricing and promotion analytics. If you are a brand and you landed here looking for a price optimization tool, that is the page you want.
What price optimization software actually does
The category divides into three lifecycle modules, and the vendors themselves divide it the same way. Revionics names Base Price, Promotions and Markdown as its three solutions; RELEX describes its own price optimization as bringing base pricing, promotions management and markdown optimization together. IDC scoped its December 2025 MarketScape on retail price optimization solutions around lifecycle price optimization, price intelligence and price analytics, and named RELEX a Leader in it. Buy one module and you have bought roughly a third of the problem, which is a reasonable thing to do on purpose and an expensive thing to do by accident.
Base price
The everyday shelf price on every item in every zone. This is the largest and least glamorous module, and it is where most of the money is, because it applies to every unit rather than to the promoted weeks. Revionics and RELEX both sell it as a distinct module from promotion and markdown.
Promotions
Which items go on deal, at what depth, in which weeks, and what the calendar returns. Retail price management software and the supplier-side promotion tools both claim this surface, and they are answering different questions with it.
Markdown and clearance
Getting seasonal, end-of-life and overstocked inventory out at the best achievable margin before a date. RELEX describes clearance optimization as choosing the discount price per item per location to reach a target by the end of a defined markdown period.
Consolidation is worth knowing about when you build a shortlist: Aptos announced its agreement to acquire Revionics on 5 August 2020 and positioned it as an extension of its merchandise lifecycle management suite. Several names on a comparison chart resolve to fewer independent companies than the chart implies.
Rule-based and elasticity-based engines
This is the single distinction that decides what you are buying, and it survives every rebrand of the category. The model families underneath are compared in more depth in price optimization models.
Rule-based engines
These encode relationships. Hold a 3 percent index to a named competitor on key value items; keep the 24-count at no more than 1.8 times the 12-count; never price a private label above 82 percent of the national brand; end every price in 9; keep two zones aligned within 5 cents. The output is a complete, defensible price file. What the engine does not have is any view of what a price change does to units.
Elasticity-based engines
These estimate a demand response per item per zone from your own history, then choose the price that maximizes a stated objective, with the rules above applied as constraints rather than as the answer. The output looks the same. The difference is that it can tell you what a recommendation is worth, and can therefore be wrong in a way you can measure.
The confusion worth avoiding
Both categories are sold as price optimization tools, and both produce a price for every item, which is what a demo shows. Ask the vendor to state the expected unit and margin effect of a single recommendation. A rule engine cannot answer that question, and a good elasticity engine answers it as a range with a confidence attached.
One rule, two items, opposite answers
A 96-store grocery chain ran a single competitive rule across its key value items: hold every one of them 3 percent under the club competitor. Two items from that list, a 12-pack of Halden sparkling water and a 12-count Voltcap energy multipack, with the elasticity each one measured on its own non-promoted weeks.
| Measure | Halden sparkling water, 12-pack | Voltcap energy, 12-count |
|---|---|---|
| Cost | $7.42 | $16.10 |
| Shelf price today | $9.99 | $24.99 |
| Club competitor price | $9.49 | $23.70 |
| Rule price, 3 percent under | $9.21 | $22.99 |
| Price change | -7.8% | -8.0% |
| Unit margin, before and after | $2.57 to $1.79 | $8.89 to $6.89 |
| Units per week, chain | 4,310 | 1,180 |
| Units needed to hold gross profit flat | 6,188 (+43.6%) | 1,523 (+29.0%) |
| Elasticity that would require | -5.6 | -3.6 |
| Measured elasticity | -1.8 | -4.6 |
| Units the measured elasticity predicts | 4,916 (+14.1%) | 1,614 (+36.8%) |
| Weekly gross profit, before and after | $11,077 to $8,800 | $10,490 to $11,120 |
| Weekly change | -$2,277 | +$630 |
The arithmetic is the whole argument. Cutting the sparkling water from $9.99 to $9.21 takes 78 cents off a $2.57 unit margin, so gross profit only holds if units rise by the ratio 2.57 divided by 1.79, which is 43.6 percent. Off a 7.8 percent price cut that needs an elasticity of -5.6. The item measures -1.8, which predicts 4,916 units, and the week ends $2,277 of gross profit lighter. The energy multipack runs the same rule and clears it: it needs -3.6 and measures -4.6, so it gains $630.
Net across just these two items, the rule costs $1,647 a week, about $85,600 a year. It is not a bad rule. It is a rule with no opinion about which items it should apply to, and on a key value item list of 300 that is a large number with no owner.
The 30-second screen, before any software is involved, is price divided by unit margin: below that elasticity no cut of any size pays for itself. The sparkling water needs 3.9 and measures 1.8; the energy multipack needs 2.8 and measures 4.6. The exact figures in the table are higher than the screen because a discrete cut also thins the margin on every unit that would have sold anyway. How to measure the elasticity in the first place is covered in price elasticity, and the index the rule is built on in retail price index.
Seven things to evaluate
Grain of the elasticity estimate
Per item per zone is the useful grain. A category-level elasticity blends a key value item against a long tail that behaves nothing like it, and then gets used to price both. Ask what happens on an item with 11 weeks of history and no price variation, because that is a large share of any real assortment.
How the rules are expressed
Price families, pack ladders, private-label gaps, ending digits, zone and banner alignment, margin floors, and the ones nobody writes down until week six. Franchise groups and multi-banner operators need a rule set that can differ by owner while holding a chain-level relationship. Get your ten hardest rules into the trial.
Cost accuracy, which is nobody's favorite topic
Every recommendation is margin arithmetic on a cost figure. If landed cost, deal cost and the invoice do not agree, the optimizer is confidently solving the wrong problem. Reconciling that is a data project you should scope before the software project.
Competitive price data
Where it comes from, how often it refreshes, and how matches are made across pack sizes and private labels. A stale or mismatched competitor file turns a competitive rule into random noise applied consistently.
Pricing simulation before you publish
You should be able to run a scenario, see the projected units, revenue and margin, and compare it against holding current price, before anything reaches a register. A tool that only produces a recommendation is asking for trust it has not yet earned.
How the price actually reaches the POS
File format, effective dating, zone mapping, batch window, failure handling, and what happens when a price is rejected at the store. This is the least discussed part of a retail price optimization software purchase and the most common reason a project stalls after go-live.
Post-hoc measurement
Whether the system reads back what happened after its own recommendation took effect, and compares it to what the model predicted. Without that loop you have a pricing opinion generator, and every retail discount management tool in the category will happily be one.
Where Scout fits
Scout reads your own POS at store and item grain and estimates base and promoted elasticity separately, per item per zone, on non-promoted weeks with distribution held constant. It holds the item file with cost and retail, so the margin arithmetic behind a recommendation runs on the same cost the invoice reconciliation uses rather than on a second copy that has drifted. It can push the resulting price file to the POS, and it reads the weeks after the change to say what the recommendation was actually worth against what the model said it would be.
The rules live alongside the model rather than instead of it: competitive index, price families, pack ladders and margin floors are constraints on the recommendation, and any item where the rule and the elasticity disagree is surfaced with the cost of the disagreement attached, which is the table above run across the whole key value item list. The pricebook side of that is in the retail pricebook entry.
One boundary, stated plainly. Scout prices; it does not buy. It does not raise or transmit purchase orders, hold an order guide, or carry an EDI connection to your suppliers, and your purchasing system keeps all three. For fuel operators there is a second line that matters more than it sounds: Scout prices the inside store only. No gallons, no fuel margin, no street-price strategy, and no forecourt data of any kind.
Related: Pricing and promotion analytics · Price optimization models · EDLP vs Hi-Lo · Pricebook audit · Retail operations software
Tell us what you’re working on
A 30-minute conversation to scope fit. Pick a time that works for you.