Research brief: Information architecture for service businesses with multiple verticals (piece 6 of 15)

Status: Research material, not finished article. Compiled May 2026.

TL;DR

• The /industries/<slug>/ URL pattern is dominant among multi-vertical B2B service firms (Crowe, BDO USA, HUB International, Dentons via /industry-sectors/, Grainger). The /for-farmers/ pattern was observed on zero of 8 sampled multi-vertical service sites — it belongs to SaaS marketing, not service business IA.
• NN/g's primary finding on audience-based navigation is a warning, not an endorsement. Audience-based nav "will often degrade usability" — works only when categories are mutually exclusive, jargon-free, with substantially unique content per section. Small service businesses rarely meet that bar.
• The dominant working model is an Industries × Services matrix with both axes in the top nav. RelaDyne, Crowe, HUB International, Dentons, E.H. Wolf all use this shape.
• Google consolidates duplicates; it doesn't penalize them. Mueller: near-duplicate vertical pages get folded together unless each has "something unique." Vertical pages without genuinely vertical-specific content will not rank.
• Multi-step forms outperform single-step only above ~7 fields. Below that, the lift evaporates (Zuko). Above that: 13.9% vs 4.5% (Formstack 2014, n=450k+); up to 300% in Venture Harbour's portfolio testing; 743% on one 11-field B2B SaaS form (Numinam).
• For small service businesses, the IA pattern that scales is hub-and-spoke: one hub page per vertical, shared services pulled in via blocks — not a full matrix of vertical × service pages. The matrix is an enterprise pattern; the hub-and-spoke is a 4-10-person-firm pattern.

How this brief decomposes

Strongest atomic claims (sources C1-C32) are filed individually. Synthesis recommendations (S1-S18) become rules where prescriptive. The decision matrix (§7) and URL pattern guide (§8) become their own reference entries since they're reusable frameworks. The Boucher & Jones implication is preserved in Boucher & Jones IA recommendation: Pattern A + hub-and-spoke, NOT enterprise matrix.

Honest gaps and caveats

• The "13.9% vs 4.5%" multi-step stat is from Formstack 2014 — a self-published account-level report across 450k+ accounts, not a controlled experiment. Industry consensus, not primary evidence. Treat as directional.
• The "300% lift" multi-step claim is widely misattributed to ConversionXL/CXL. Correct primary attribution: Venture Harbour (Marcus Taylor) portfolio testing.
• The "~12 item mega-menu trigger threshold" attributed to NN/g in earlier drafts does not exist in NN/g's primary literature. The real NN/g guidance is qualitative: mega menus are appropriate "for bigger sites with many features."
• NN/g and Baymard have no dedicated article on the "industries we serve" pattern specifically. Closest is the audience-based-nav warning, which addresses adjacent but not identical territory.
• The Gartner six-jobs framework + 77% / 6-10 stakeholders / 95% revisit figures all trace to a single 2019 CSO Update report. Widely cited; treat as Gartner-attributed industry consensus.
• The polar bear book (Rosenfeld/Morville/Arango 4th ed., 2015) is foundational but pre-mobile-first-indexing. Principles current; examples dated.
• Schema.org best practice for multi-audience services has limited primary documentation beyond type definitions. Practitioner blogs (Aubrey Yung) fill the gap as single-source.
• Mobile navigation patterns for multi-vertical sites are underexplored in current research. NN/g mobile guidance applies, but no 2024-2026 study tests multi-vertical service-business mobile nav specifically.

Editorial direction

1. Lead with the URL convention finding (/industries/<slug>/ is the dominant pattern) — it's the most concrete, immediately actionable claim.
2. Treat the Decision Matrix (§7 → Reference: which IA pattern to use, by business shape (decision matrix)) as the deliverable readers save.
3. NN/g's audience-nav warning is the contrarian hook — most agency writing treats audience-nav as a best practice; the actual research says the opposite.
4. The hub-and-spoke vs matrix distinction is the small-business-vs-enterprise honesty. Don't pretend a 4-person firm should build Dentons IA.