Methodology

How scores are computed, what the data actually measures, and where the gaps are.

Opportunity score formula

Every HS-6 code receives a single opportunity score. Higher scores indicate codes where Vietnam has proven export capability, Canada has meaningful import demand, the buyer field is open, and (optionally) the HS-4 heading is present in a configured catalog.

Codes below $1M USD in Vietnam world exports are not scored — supply is too thin to form a reliable signal.

What each factor measures — and its limits

Supply scale (log₁₀)

Vietnam's total world export value for the HS-6in the latest Comtrade year (2024). Log-transformed so a $10B exporter doesn't automatically outrank a $100M one by 100×. Without the log, the score would be almost entirely determined by export size and the other factors would matter little.

Supply gap

1 minus Vietnam's current share of Canadian imports for this code. A gap of 0.95 means Vietnam holds ~5% of what Canada currently imports — 95% of the market is served by others. A gap of 0.10 means Vietnam is already the dominant supplier.

Fragmentation (1 − top-3 share)

Derived from CID data. A fragmentation score of 0.79 means the top-3 Canadian importers control 21% of import value — the field is open and no single buyer dominates. A score of 0.20 means top-3 controls 80% — entering means either displacing incumbents or surviving on what remains.

This is a pure concentration measure. Importer count is not part of this factor — see the separate section on importer count below.

Catalog-adjacency bonus (×1.25)

A 1.25× multiplier (not additive) applied when the HS-4 heading is present in a deployment-configured or user-supplied catalog of HS-4 codes. The logic: existing supplier relationships, quality systems, and customer knowledge lower the barrier relative to a cold entry. When no catalog is configured, catalog_bonus is 1.0 for all codes and the factor is neutral — scores are unaffected.

What CID tells us — and doesn't

The CID provides the importer concentration data used in this system. It covers Canadian businesses that imported goods with a customs value ≥ C$1,000 in a given year.

What CID provides

For each HS-6: total import value in CAD, total importer count, percentage share held by the top 3, top 10, and top 25 importers. For importers above ISED's disclosure threshold, company names are also available.

What CID does NOT provide

Individual company-level shares for suppressed importers. We know the top-3 together hold X% but not how that is distributed. A 21% top-3 share could mean one player at 19% and two at 1%, or three roughly equal players at 7% each — these have very different entry implications. Use the tier table (top-3 / top-10 / top-25) to reason about the distribution.

CID also does not identify the country of origin for those imports. A high total CAD value for an HS-6 includes imports from all countries, not just Vietnam.

What "fewer importers" really means

Importer count and concentration are independent dimensions. The earlier version of this score conflated them — an understandable error but a meaningful one.

Fewer importers = smaller market. A market with 8 importers totalling $12M CAD is small in absolute terms — limited headroom, limited revenue potential. But those 8 importers could each hold ~12.5% (perfectly fragmented) with no dominant player. Entry is achievable; the constraint is market size.

Many importers ≠ hard to enter. A market with 80 importers where top-3 hold 78% is large but oligopolistic. Entry means displacing or differentiating against the dominant trio.

The right question: is there a dominant incumbent I must displace? — answered by top-3 concentration (fragmentation factor). How big is the prize? — answered by importer count and total CAD value. This system now surfaces both signals separately rather than blending them into one score component.

Year asymmetry between data sources

The primary trade flow data is Comtrade 2024 (Canada-Vietnam bilateral). The CID concentration data is 2023 — one year behind. This creates a structural asymmetry: the supply gap and supply scale reflect 2024 trade, while fragmentation and market size reflect 2023 buyer structure.

In stable markets this is unlikely to matter. In fast-moving codes (commodities with volatile prices, electronics with rapid SKU turnover), the 2023 concentration picture may be stale. Validation flags surface codes where year discrepancy is combined with other signals of instability (large YoY swings, mirror-trade gaps).

What the score does NOT capture

The score is a signal filter, not a business case. It intentionally excludes factors that require primary research:

• Tariffs — already mostly CPTPP Free for Vietnam. The MFN baseline is shown per code for awareness. Tariff advantage is not included in the score because it's nearly uniform across codes.
• Shelf life / cold chain — fresh produce has logistical constraints the score ignores. High-scoring fresh codes may be logistically impractical without established airfreight.
• Channel access — retail, foodservice, and industrial buyers have different onboarding requirements. The score treats all importers equally.
• Brand vs. commodity positioning — some high-score codes are branded categories where private-label entry is hard. Others are pure commodity where price is the only lever.
• Internal demand signals — customer RFQ data, rep feedback, and pipeline data are not integrated. The score is a market filter; sales pipeline context should overlay it.
• BEC5 end-use classification — whether a code is an intermediate good, capital good, or consumer good affects the sales motion. Not yet incorporated.

Data sources

Eight primary sources feed this system. Each is pulled programmatically and stored in a local SQLite database. Refresh cadences reflect each source's release schedule.

Trade flow values from different reporters (Canada importer vs Vietnam exporter) will not match perfectly due to CIF/FOB valuation differences and classification discrepancies. The mirror trade gap is surfaced as a validation flag where it exceeds expected ranges.