Files
2026-07-13 12:35:03 +08:00

11 KiB
Raw Permalink Blame History

Competitor Landscape Guide

Guide Origin: Official | ArcKit Version: [VERSION]

/arckit:competitors discovers competitor landscape intelligence — rival suppliers, awarded-value market share, head-to-head comparison, and concentration — from live UK procurement notices via the UK Tenders MCP.

Agent Architecture: This command uses a three-tier reader/writer subagent split. The arckit-tenders-reader subagent fetches evidence from the UK Tenders MCP (shared with /arckit:tenders, keeping that I/O isolated from your main context window). The orchestrator tier (the slash command) validates the reader's JSON payload against the tenders-handoff.schema.json schema and computes deterministic derived fields, including the supplier-rivalry head-to-head. The arckit-competitors-writer subagent then renders the final artefact. The slash command launches and coordinates both agents.


What is Competitor Landscape Intelligence?

Competitor landscape intelligence is the systematic analysis of real awarded-contract data to understand who else is winning work in a capability space — their share of awarded value, which buyers they serve, and how a focal supplier compares against each rival.

Key questions answered:

  • Who are the rival suppliers in this capability space, and what is their awarded-value share?
  • How does a named supplier compare against each rival (shared buyers, recent wins)?
  • How concentrated is the competitive landscape — is one supplier dominant?
  • Which government buyers are active in this space?

Mandatory caveat. Awarded value is not actual spend; figures are for market context and benchmarking, not the costed Economic Case. Never use CMPT figures as the cost line in a Green Book / Orange Book Economic Case without independent substantiation.


When to Use

  • Before a vendor evaluation — understand who has already won similar work and at what scale
  • Before scoping a competition — know the realistic supplier pool and whether concentration is a risk
  • When building a Company Experience scorecard — rival award history provides comparative context for /arckit:score
  • When assessing a sole-source or incumbent renewal — evidence the broader competitive landscape
  • As part of a build-vs-buy analysis — quantify the commercial supply depth alongside /arckit:research

Prerequisites and Dependencies

Artefact Dependency Why
Requirements (ARC-*-REQ-*.md) OPTIONAL Capability keywords and CPV codes can be derived from DR/FR/INT requirements
Architecture Principles (ARC-000-PRIN-*.md) OPTIONAL Identifies the organisational context for capability scoping
Vendor Profiles (vendors/{slug}-profile.md) ENRICHED Pre-existing vendor profiles gain an enriched Government Award History section
Risk Register (ARC-*-RISK-*.md) DOWNSTREAM Supplier-concentration findings feed R-xxx concentration/single-supplier-dependency risks

Inputs

The command accepts three input forms, which can be combined:

Input Flag / Form Example
Supplier focus (focal supplier) --supplier 'Name' /arckit:competitors --supplier 'Acme Cloud Ltd'
Capability (free text) bare arguments /arckit:competitors cloud hosting
CPV code --cpv NNNNNNNN /arckit:competitors --cpv 72200000

Use --supplier 'Name' to run a supplier-focus query: the named supplier becomes the focal point and the command produces a head-to-head comparison against each rival. Without --supplier, the command runs in capability-focus mode: it returns the full competitive set ranked by awarded-value share, with no focal supplier or head-to-head.

You can combine flags: /arckit:competitors cloud hosting --cpv 72200000 --supplier 'Acme Cloud Ltd'.

CPV code format: eight digits, optionally followed by a division suffix (NNNNNNNN-N). Common codes:

Code Category
72200000 Software-related services
72300000 Data services
72600000 IT support and consultancy
48000000 Software packages and systems
79000000 Business services
73000000 R&D services

Scenario Matrix

Scenario Prompt Focus
Capability landscape /arckit:competitors digital identity verification Rival suppliers ranked by awarded-value share
Supplier head-to-head /arckit:competitors --supplier 'Acme Cloud Ltd' Focal supplier vs each rival
Scoped supplier analysis /arckit:competitors --cpv 72600000 --supplier 'Acme Cloud Ltd' Focal supplier vs rivals in a specific CPV
Market concentration check /arckit:competitors cloud hosting --cpv 72200000 Concentration flag and supply-base depth
Vendor evaluation context /arckit:competitors managed security operations Who is winning, at what value, for which buyers

Command

/arckit:competitors [project-number-or-name] <--supplier 'Name' | capability | --cpv NNNNNNNN>

Outputs: projects/<id>/research/ARC-<id>-CMPT-<NNN>-v1.0.md

Multi-instance type: CMPT artefacts are sequenced within the project's research/ directory. Re-running produces a new numbered artefact (CMPT-001, CMPT-002, …) rather than overwriting the previous run.


Long runs: Remote Control + push notifications

/arckit:competitors can take several minutes when the UK Tenders MCP needs to page through large result sets. To avoid babysitting the terminal, pair it with Claude Code Remote Control:

claude remote-control

Drive the session from claude.ai/code or the mobile app, then enable /config → Push when Claude decides so your phone gets a notification on completion. ArcKit's minimum Claude Code floor (v2.1.121) already covers the v2.1.110 RC requirement.

Caveats: Pro/Max plans only (no API keys, no Bedrock/Vertex/Foundry), push is a single on/off so chatty agents can over-notify, and the local claude process must keep running.


Artefact Sections

A completed CMPT artefact contains:

  1. Executive Summary — 35 key findings, concentration flag, focal supplier position (supplier-focus runs only).
  2. Query Scope — recorded focus, CPV, supplier, keywords, date range so the run is reproducible.
  3. Competitive Set — rival suppliers ranked by awarded-value share %, with award counts and key buyers.
  4. Focal Supplier — on supplier-focus runs, the named supplier's awarded value, award count, and share pulled out of the competitive set.
  5. Head-to-Head — on supplier-focus runs, the focal supplier compared against each rival: shared buyers, rival's most recent win. Renders not-applicable on capability-focus runs.
  6. Per-Rival Buyer Relationships and Recent Wins — short prose per rival derived from the data.
  7. Concentration — top-1 / top-3 share and a HIGH / MEDIUM / LOW flag with the rule applied.
  8. Representative Notices — sample award notices with their notice_url (required for OGL attribution).
  9. Data Freshness and Source HealthData current as of <ISO datetime> or a freshness-unavailable note when the status endpoint was unreachable.
  10. Caveats — including the mandatory awarded-value caveat.

Concentration Flag Rules

Flag Condition
HIGH Top-1 supplier share > 50% or top-3 supplier share > 80%
MEDIUM Top-3 supplier share > 60% (and HIGH conditions not met)
LOW All other cases

A HIGH concentration flag should be registered as a supplier-dependency risk in the project Risk Register (/arckit:risk).


Vendor Profile Enrichment

When a project already has a vendor profile (vendors/{slug}-profile.md) for a rival supplier, the arckit-competitors-writer subagent enriches that profile's Government Award History section with the award data from the competitive set. This means running /arckit:competitors before /arckit:score gives the score command richer comparative evidence to work with.

The writer performs a bounded section-merge — it updates only the ## Government Award History section of pre-existing profiles. It does not create new profile files.


Data Source

Property Detail
MCP server uk-tenders (keyless, deferred load, shared with /arckit:tenders)
Coverage ~677,000 UK contracting processes across five national portals
Portals covered Find a Tender Service (FTS), Contracts Finder, Public Contracts Scotland, Sell2Wales, eTendersNI
Refresh cadence Nightly
Availability Best-effort single Cloud Run instance — no formal SLA. A dead endpoint degrades gracefully (the reader returns partial data and the artefact renders with degraded-source notes).
Licence Open Government Licence v3.0 (OGL v3.0)
Attribution Every notice record carries its official notice_url; artefacts must include the OGL v3.0 attribution line

OGL Attribution

Every CMPT artefact includes the attribution line:

Contains public sector information licensed under the Open Government Licence v3.0.

The source data is re-published by the UK Tenders MCP from official UK procurement portals verbatim under the OGL v3.0. Every supplier figure and share percentage in the artefact traces to a notice_url citation from the relevant portal.


Evaluation Criteria Explained

The reader returns evidence across five procurement portals, ranked internally by:

Criterion Purpose
Awarded value Primary ranking signal for rivals and share calculations
Award count Proxy for market depth (many small awards vs few large awards)
Share % Concentration measurement; feeds the top-1 / top-3 concentration flag
Date range coverage Assesses recency and whether the competitive landscape has changed materially
Source health Flags which portal feeds were degraded at fetch time

Integration with Other Commands

Direction Command Integration
Input /arckit:requirements CPV codes and capability keywords derived from DR/FR/INT
Input /arckit:tenders TNDR artefact narrows the capability scope and provides market benchmarks
Output /arckit:research Feed the competitive set into build-vs-buy analysis and TCO comparison
Output /arckit:score Rival award history used as comparative Company Experience evidence
Output /arckit:risk Concentration / single-supplier-dependency risk
Output /arckit:adr Competitive landscape and route-to-market decisions recorded as ADRs

Follow-on Actions

  • Feed the competitive set into build-vs-buy analysis (/arckit:research)
  • Use rival award history as Company Experience evidence in a vendor scorecard (/arckit:score)
  • Register supplier-concentration or single-supplier-dependency risks (/arckit:risk)
  • Record route-to-market decisions as ADRs (/arckit:adr)
  • Escalate HIGH concentration findings to commercial lead before procurement scoping