What CRMA is
CRMA stands for Competitive Retailer Marketing Area, the geography made up of a retailer's own stores plus the competing stores in the same counties. The CPG Data Tip Sheet defines it as "IRI's term for the overall geographic area in which a retailer operates," used "to compare performance of a retailer and its competition." When a buyer at a Southeast grocer opens a category review and asks how your brand does in their market rather than in their stores, CRMA is the geography that answers.
Almost every analyst learns channel definitions first and geography never. xAOC, MULO and Natural get explained on day one, and all three are channels: which retailer types are in the aggregate. CRMA is a different axis. It is which physical counties are in the aggregate, holding the retailer fixed. xAOC and MULO will not answer a question about one banner versus its neighbours; CRMA will.
TA, CRMA and ROM: the trio
Three geographies travel together, and mixing them up is the single most common way a category review number gets disputed.
| Level | What it contains | The question it answers |
|---|---|---|
| TA | The retailer's own stores only | How am I doing inside this banner? |
| CRMA | The retailer plus competitors in its counties | How is the whole market doing? |
| ROM | CRMA minus the retailer | How am I doing everywhere except this banner? |
TA is the Trading Area, the retailer's own store set. ROM is the Remaining Market, or Rest of Market, and the CPG Data Tip Sheet describes it as the piece that "gives you a single comparison point for each retailer, a weighted average of all the competitors."
The arithmetic that matters: TA plus ROM equals CRMA in dollars and units, but not in share. Shares are ratios, so they weight by the size of each geography rather than adding. Analysts who add two share numbers and get a third get caught in the room.
One methodology detail worth knowing before you quote a CRMA number as though it were census. Circana's InfoSource documentation is explicit that an RMA, the single-retailer geography, "uses census data (all scans/all stores) for a single retailer chain," while a CRMA "includes census data for a single retailer and a combination of census data and sampled data for that retailer's regionally applicable competitors." The retailer half is measured. The competitive half is partly projected. Treat a two-tenths-of-a-point share move in ROM with more caution than the same move in TA.
Worked example: one banner against its own market
A salty snacks brand, 52 weeks, one Southeast grocery banner and the CRMA around it.
| Geography | Brand dollars | Category dollars | Brand share |
|---|---|---|---|
| TA (the banner) | $2,410,000 | $63,400,000 | 3.80% |
| ROM (everyone else) | $5,530,000 | $242,000,000 | 2.29% |
| CRMA (both) | $7,940,000 | $305,400,000 | 2.60% |
Check the additivity: $2,410,000 + $5,530,000 = $7,940,000, and $63,400,000 + $242,000,000 = $305,400,000. Now check the shares, which do not add: 3.80% and 2.29% do not sum to 2.60%, they weight to it, because ROM is roughly four times the size of TA.
The read is one number. The brand holds 3.80% share inside the banner against 2.29% in the surrounding market, a development index of 166. It is 66% better developed with this retailer than with its neighbours.
That single index reframes the meeting. A buyer looking at a 3.80% share sees a small brand. A buyer looking at an index of 166 sees a brand that performs better in their stores than in anyone else's, which is an argument for distribution rather than a reason to defend it. Run the same index the other way and it is equally useful: an index of 60 says the banner is under-supporting an item the market likes, and the cause is usually distribution or shelf position rather than demand.
Pair the index with velocity before drawing the conclusion. A high development index built on twice the distribution is a distribution story; the same index at equal distribution is a genuine velocity story, and only the second one earns more shelf.
Where CRMA breaks down
It is a county-based construction, which produces two predictable failures.
Retailers whose stores cluster in dense metros get a CRMA that includes counties where they have no presence at all, so ROM contains competitors they never actually meet. And a banner operating across many states gets one CRMA that averages a strong region with a weak one, which is exactly the pattern that makes a national number look flat while both halves are moving. Where the retailer publishes regional aggregates, use those instead, or split the CRMA manually.
The other limit is channel coverage inside the geography. A CRMA is only as complete as the panel behind it, so club and hard-discount competition inside those counties is usually thin or absent. A ROM that excludes the discounter taking your volume will show your brand holding share while the category quietly migrates.
Where Scout fits
Scout reads the TA-versus-ROM comparison off the syndicated files you already license, alongside the retailer POS you receive directly, and computes development indices by brand, item and category without the geography being rebuilt in a spreadsheet each time. It does not define or sell the geography: CRMA, TA and RMA are Circana constructions, and Scout reads them rather than producing them. For retailers on their own POS, the same comparison runs store-to-store against Scout's own data, which is a different and narrower benchmark than a syndicated CRMA.
The short version
- CRMA is the Competitive Retailer Marketing Area: a retailer's stores plus competing stores in the same counties, used to benchmark a banner against its own market.
- TA is the retailer's own stores. ROM is CRMA minus the retailer. Dollars and units add across the trio; shares do not, they weight.
- The retailer half of a CRMA is census data. The competitive half is partly sampled and projected, so read small ROM moves with more caution.
- Share in TA divided by share in ROM is the development index. The worked example ran 3.80% against 2.29%, an index of 166, which is a distribution argument rather than a defence.
- County construction misfires for metro-clustered and multi-region banners, and a thin club and discount panel inside the geography will hide the competitor actually taking the volume.
Sources: CPG Data Tip Sheet, "Competitive Retailer Marketing Area (CRMA)"; CPG Data Tip Sheet, "Remaining Market (ROM)"; Circana InfoSource, "What Are the Differences in Retailer Composition for RMAs, CRMAs and the Region/State/Market Geographies?".