Skip to content

See a demo

30 minutes with Sasha Zhang · video link on confirmation

Loading scheduler…

CPG glossary

Retail competitive intelligence, and its blind spots

What retail competitive intelligence is

Retail competitive intelligence is the practice of measuring what competing brands and banners are actually doing, from transaction data rather than from inference. It answers four questions with numbers: what share they hold, how fast their items sell, how many doors they are in, and what shoppers pay. Everything else in the discipline is commentary on those four.

That framing is deliberately narrow, because the search results for this topic are dominated by SWOT templates, and a SWOT produces adjectives. "Strong brand equity, weak digital presence" is not intelligence. A competitor at 4.2% share growing 18% on 340 more doors while holding price is intelligence, because every clause in it can be checked and none of it needs a guess about their strategy.

The four numbers, and where each one comes from

NumberThe questionBest source
ShareHow much of the category do they hold?Syndicated scan, or portal category totals
VelocityHow hard does each item sell per door?Syndicated scan, or your own retailer POS
DistributionHow many doors and how much ACV?Syndicated scan (TDP and % ACV)
Price gapWhat do shoppers pay, versus your shelf?Retailer POS, or shelf checks

Share on its own is the number that misleads, because two very different things produce it. A competitor gaining share on distribution is a sales-team story you counter at the buyer's desk. A competitor gaining share on velocity at flat distribution is a product story you cannot counter with a sales call. The velocity, share and TDP decision tree sets out which one to lead with, and TDP is the distribution measure that keeps the two apart.

What each source cannot see

This is the part generic treatments skip, and it is the part that decides whether a competitive read is trustworthy.

SourceWhat it genuinely showsWhat it cannot seeTypical lag
Syndicated scan (Circana, NielsenIQ, SPINS)Competitor dollars, units, distribution, average price, promoted shareAnything outside the panel; club and hard discount are thin; their cost or margin1 to 3 weeks
Your own retailer POSEvery competing item in your stores, at store-week grainTheir performance in stores you do not operate1 to 7 days
Retailer portalsYour items plus category totals at that retailerNamed competitor detail, which is usually masked into an aggregate1 to 3 days
Shelf and price checksActual shelf price, facings, signage, out-of-stocks on one dayVolume; and it does not scale without a field teamSame day
Public filings and earnings callsSegment revenue, unit growth, stated strategy, capexAnything at item or store grain, and it is heavily framed4 to 8 weeks

The pattern is worth naming. The sources that are fast and cheap are narrow, and the sources that are broad are slow and partly projected. Nothing gives you a competitor's item-level performance in stores you do not operate at a speed that supports a weekly decision. A retail competitive intelligence programme that promises otherwise is selling inference as measurement.

The subtraction trick

You can measure the competition without any competitor's data file, and this is the read most brands already have available and do not run.

Retailer portals give you your own items plus the category total. Subtract the first from the second and you have everything else in the category, at that retailer, at whatever grain the portal reports.

Refrigerated salsa at one grocery banner, 13 weeks against the same weeks a year ago:

LineYear agoThis yearChange
Category total$1,704,000$1,842,000+8.1%
Your brand$241,900$286,500+18.4%
Everything else$1,462,100$1,555,500+6.4%
Your share14.20%15.55%+1.35 pts

Check it: $1,704,000 minus $241,900 is $1,462,100, and $1,842,000 minus $286,500 is $1,555,500. Your brand grew 18.4% while the rest of the category grew 6.4%, which is 12 points of outperformance and 1.35 points of share, and none of it required a competitor's file.

What the subtraction cannot do is name anyone. "Everything else" is one number covering fourteen brands, so it tells you whether you are winning without telling you who you are beating. Pair it with syndicated data once a quarter to put names on the movement, and use the portal weekly to know the direction before the syndicated file lands.

Sizing a competitive landscape honestly

A competitive landscape analysis is usually presented as a two-by-two with logos on it. The version that survives a buyer meeting is a list with a cut line: every brand above a distribution or share threshold in the categories you compete in, sized by the same four numbers, refreshed on a fixed cadence.

Three rules keep it from rotting:

  • Set the boundary before you look. Define the category the way your retailer defines it, not the way that flatters you. A landscape drawn around your own subsegment will always show you leading.
  • Include private label as a competitor. It is frequently the largest share holder in the set and the one most often excluded from the chart, and private label behaves differently from national brands under promotion.
  • Refresh on a schedule, not on a scare. A landscape rebuilt only when someone launches gives you a snapshot with no trend, and trend is where the finding lives.

For the price half of the discipline specifically, the retail price index goes deeper than this page does.

Where Scout fits

Scout reads the syndicated files and retailer POS you already license and keeps the four numbers current by brand, item and retailer, including the subtraction read above, so a competitive review is a page you open rather than a spreadsheet somebody rebuilds. The boundary: Scout reads transactions. It does not scrape competitor sites, recognise shelf images, or observe shopper foot traffic, so its competitive visibility is exactly as wide as the data you feed it and no wider. If you want the commercial version of this rather than the method, the retail intelligence platform page covers what that looks like in product.

The short version

  • Retail competitive intelligence is four measurable numbers: share, velocity, distribution and price gap. A SWOT produces adjectives instead, which is why it does not survive a buyer meeting.
  • Every source has a blind spot. Syndicated data is broad, slow and partly projected; your POS is fast and narrow; portals mask competitor names; filings are quarterly and framed.
  • Category total minus your own items gives you the rest of the category for free. In the worked example that was +18.4% against +6.4%, worth 1.35 share points.
  • The subtraction cannot name anyone. Use syndicated data quarterly for names and the portal weekly for direction.
  • Draw the landscape boundary the way the retailer draws it, include private label, and refresh on a schedule so you get a trend rather than a snapshot.
See your CPG data answer questions in plain English — book a Scout demo

Want the rest of the CPG analyst's glossary?

Drop your email and we'll send the full set of CPG and retail-data definitions as one reference sheet.