Look-Alike Account Finder
From a handful of accounts you wish you had more of, surface a ranked look-alike cohort with the firmographic and technographic patterns that placed each account on the list.
Look-Alike Account Finder: Siemens · Bosch · ABB
Source: HG Insights, May 2026. All firmographic figures are HG-observed (high confidence); technographic intensity and install signals are HG-observed-and-modeled.
Pattern Signature
The three seeds converge on five core features. Any candidate matching three or more qualifies; four or five is a strong match.
| # | Feature | Siemens | Bosch | ABB | Basis |
|---|---|---|---|---|---|
| 1 | NAICS-4: 3344 — Computer & Electronic Product Manufacturing | ✅ (334513) | ⚠️ adjacent (336320) | ✅ (334419) | HG firmographic |
| 2 | Employees 100K–500K | 316K | 418K | 112K | HG firmographic |
| 3 | Revenue $10B–$110B | $88.7B | $102.3B | $33.2B | HG firmographic |
| 4 | EMEA HQ (Germany / Switzerland) | ✅ Germany | ✅ Germany | ✅ Switzerland | HG firmographic |
| 5 | SAP (ERP) — high-intensity, 100+ locations across seeds | intensity 7,463 | intensity 5,809 | intensity 1,550 | HG technographic |
| 6 | Engineering toolchain — Autodesk/AutoCAD + Dassault Systèmes + MathWorks | ✅✅✅ | ✅✅✅ | ✅✅✅ | HG technographic |
Signature applied: EMEA large-cap industrial manufacturer ($10B+ revenue, 50K+ employees), SAP at scale, plus ≥2 engineering design tools (Autodesk, Dassault Systèmes, MathWorks).
Look-Alike Table (Top 15)
Ranked by number of pattern features matched (5 max), then revenue scale proximity to seed median (~$75B centroid).
| # | Company | Domain | Country | Revenue | Employees | Industry (NAICS-4) | SAP | Autodesk/CAD | Dassault Sys | MathWorks | Score / 5 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | ThyssenKrupp AG | thyssenkrupp.com | Germany 🇩🇪 | $36.9B | 93,375 | Manufacturing (3344 adj.) | ✅ intensity 5,415 | ✅ intensity 2,702 | ✅ intensity 2,438 | ✅ intensity 691 | 5/5 |
| 2 | Airbus SE | airbus.com | Netherlands 🇳🇱 | $82.5B | 165,294 | Manufacturing | ✅ intensity 3,768 | ✅ intensity 1,126 | ✅ intensity 2,388 | ✅ intensity 1,436 | 5/5 |
| 3 | Koninklijke Philips N.V. | philips.com | Netherlands 🇳🇱 | $20.0B | 65,340 | Manufacturing | ✅ intensity 7,051 | ✅ intensity 2,704 | ✅ intensity 3,686 | ✅ intensity 3,774 | 5/5 |
| 4 | Hitachi, Ltd. | hitachi.com | Japan 🇯🇵 | $65.1B | 268,660 | Manufacturing | ✅ intensity 3,782 | ✅ intensity 1,698 | ✅ intensity 1,502 | ✅ intensity 1,462 | 4/5 (geo miss) |
| 5 | AB Volvo | volvogroup.com | Sweden 🇸🇪 | $48.6B | 98,844 | Manufacturing | ✅ intensity 3,832 | ✅ intensity 2,641 | ✅ intensity 2,342 | ✅ intensity 2,041 | 5/5 |
| 6 | Safran | safran-group.com | France 🇫🇷 | $29.6B | 99,364 | Manufacturing | ✅ intensity 1,869 | ✅ intensity 1,898 | ✅ intensity 2,435 | ✅ intensity 1,768 | 5/5 |
| 7 | Nokia Oyj |
⚠️ ABB is excluded as a seed. No other explicit exclusions were provided.
Score Legend & Pattern Citations
Each score component maps directly to a verified HG signal or search filter:
- SAP ✅ = company_technographic confirmed SAP/SAP ERP at medium-to-high intensity (>500), verified within 12 months
- Autodesk/CAD ✅ = Autodesk or AutoCAD verified install
- Dassault Systèmes ✅ = Dassault Systèmes PLM/CAD install confirmed
- MathWorks ✅ = MathWorks / MATLAB confirmed install (engineering simulation anchor)
- EMEA ✅ = company HQ in Germany, Switzerland, France, Netherlands, Sweden, Finland, or UK
- Revenue / Employees / Industry / Country = company_firmographic enrichment of each candidate (search_companies returns identity only; these columns are pulled per-candidate via company_firmographic, high confidence)
Top-5 Deep Profiles
🥇 ThyssenKrupp AG — Score 5/5
Domain: thyssenkrupp.com | Country: Germany | Revenue: $36.9B | Employees: ~93K
- SAP: intensity 5,415 (110 US locations, 158 DE locations, verified May 2026) — deepest SAP penetration after Siemens/Bosch
- Autodesk/AutoCAD: intensity 2,702 / 2,619 (34 + 31 locations, fresh)
- Dassault Systèmes: intensity 2,438 (27 US locations)
- MathWorks: intensity 691 / 688 (6 locations, verified May 2026 — note: stale Jun 2025 on US signal; treat as medium)
- Why it ranks #1: German HQ, large-cap industrial manufacturer, all 5 signature tech anchors verified and fresh, EMEA-native, revenue within 1 band of seed midpoint.
🥈 Airbus SE — Score 5/5
Domain: airbus.com | Country: Netherlands (HQ) | Revenue: $82.5B | Employees: ~165K
- SAP: intensity 3,768 (87 US, 51 FR, 52 DE locations, verified May 2026)
- Dassault Systèmes: intensity 2,388 (41 US, 45 FR, 24 DE, 119 CA locations — exceptionally broad)
- Autodesk/AutoCAD: intensity 1,126 / 1,069 (22 US, verified April 2026)
- MathWorks: intensity 1,436 (25 FR locations, verified April 2026)
- Why it ranks #2: Revenue closest to Siemens ($82.5B vs $88.7B), all 5 anchors confirmed at scale across multiple geographies, pure EMEA/aerospace manufacturer with identical engineering toolchain.
🥉 Koninklijke Philips N.V. — Score 5/5
Domain: philips.com | Country: Netherlands | Revenue: $20.0B | Employees: ~65K
- SAP: intensity 7,051 (147 US locations, verified May 2026) — highest SAP intensity in the pool, matching Siemens
- MathWorks/MATLAB: intensity 3,774 / 3,762 — strongest MATLAB signal outside the seeds
- Dassault Systèmes: intensity 3,686 (38 US locations, verified May 2026)
- Autodesk/AutoCAD: intensity 2,704 / 2,315
- Why it ranks #3: All 5 anchors at high intensity (>70 band), EMEA HQ, NAICS-4 aligned with Siemens and ABB. Revenue is lower ($20B), but the technographic overlap is the strongest in the cohort.
4️⃣ Hitachi, Ltd. — Score 4/5
Domain: hitachi.com | Country: Japan | Revenue: $65.1B | Employees: ~269K
- SAP: intensity 3,782 (US) + 1,707 (India) — global footprint
- Autodesk/AutoCAD: intensity 1,698 / 1,578
- Dassault Systèmes: intensity 1,502
- MathWorks: intensity 1,462
- Geography penalty: Japan HQ — APAC, not EMEA. Adjacent employee/revenue band, but the geo axis is a miss.
5️⃣ AB Volvo — Score 5/5
Domain: volvogroup.com | Country: Sweden | Revenue: $48.6B | Employees: ~99K
- SAP: intensity 3,832 (40 US, 20 IN, 12 SE, 21 FR locations)
- Autodesk/AutoCAD: intensity 2,641 / 2,590 (29 + 28 locations, fresh)
- Dassault Systèmes: intensity 2,342 (25 US locations)
- MathWorks/MATLAB: intensity 2,041 / 1,969 (11 US locations)
- Why it ranks #5: EMEA-HQ (Sweden), 5/5 anchors, revenue mid-band, large EMEA industrial manufacturer. Slight geo note: primarily automotive/transport vs. process control/electrification, but technographic fingerprint is nearly identical.
Confidence Note
Signature strength: Strong (4–5 features well-populated). SAP + engineering CAD (Autodesk, Dassault, MathWorks) is a tight, defensible signature — few companies outside heavy-industry manufacturing carry all three. The main uncertainty is:
- Bosch NAICS divergence. Bosch sits in NAICS 336320 (Motor Vehicle Electrical), not 3344. The signature was broadened to "large EMEA industrial manufacturer" rather than pure NAICS-4 = 3344. This was the right call — Bosch's tech stack and employee band are perfectly aligned; the NAICS code is a known coarse-resolution issue.
- Scores 3/5 (ranks 8–15) have SAP + Autodesk confirmed via search filter but Dassault Systèmes and MathWorks were not individually verified by technographic pull — treat those as directional matches, not confirmed.
- Hitachi is the strongest non-EMEA match: if the use case is global (not EMEA-specific), it would rank #2 or #3.
- Atlas Copco matches 5/5 but sits one revenue band below ($17B vs. the $33B–$103B seed range). Its technology install intensities are lower (sparse install signal), though verified fresh. Best suited for mid-market peer comparison, not a direct Siemens/Bosch analog.
Overview
Find accounts that resemble your best customers. Marketing Ops feeds a seed list of high-fit accounts; the workflow extracts the firmographic and technographic patterns those seeds share, runs `search_companies` against those patterns, and returns a ranked look-alike cohort with similarity scores and the signals that placed each account on the list.
Use cases
Build the next ABM target list from the accounts you already love
Feed in five to ten best-fit accounts. The workflow extracts the shared firmographic + technographic signature, runs it through `search_companies`, and returns a ranked look-alike cohort. Each row cites the two to three patterns that placed the account on the list, so SDR leadership can defend why these and not others.
View workflow prompt
# Look-Alike Account Finder
## Parameters
- `{{seed_domains}}` *(required)* — Comma-separated domains of the seed accounts. Example: `siemens.com,bosch.com,continental.com`
- `{{cohort_size}}` *(optional)* — How many look-alikes to return. Example: `25`
- `{{exclude_domains}}` *(optional)* — Comma-separated domains to exclude from results. Example: `abb.com,emerson.com`
## Purpose
Take {{seed_domains}}, extract the firmographic and technographic patterns those seeds share, return a ranked look-alike cohort of {{cohort_size}}. Exclude any in {{exclude_domains}}. Every result cites the patterns that placed it on the list.
## Process
1. **Profile the seeds.** `company_firmographic` and `company_technographic` (intensity-sorted, 30 products) across {{seed_domains}}. Capture shared industry codes, revenue and employee bands, geo concentration, top-5 high-intensity products, top-3 categories.
2. **Build a pattern signature.** Synthesize the three to five strongest shared features: industry tag plus revenue band plus geo plus 2 to 3 technographic anchors. State the signature explicitly so the reader can challenge it.
3. **Search the universe.** `search_companies` against the signature: industry + revenue band + geo + each technographic anchor. Run two to four searches with progressively looser constraints; merge and dedupe. `search_companies` returns identity only (companyId/companyName/domain) — the revenue/employee/geo filters scope the result but those values are not in the output.
4. **Enrich the candidates.** For the merged candidate set, call `company_firmographic` (pass each returned companyId as hg_id, or the domain) to pull revenue, employees, industry, and country — the firmographic columns that fill the look-alike table.
5. **Score similarity.** For each candidate: count how many of the seed pattern features match (firmographic from step 4 + technographic). Higher matches rank higher. Cap at {{cohort_size}}; exclude {{exclude_domains}}.
6. **Surface the table.** Each row: company, domain, industry, revenue band, key matching signals (the 2 to 3 patterns it scored on), similarity score.
## Output Format
- Pattern signature (3 to 5 features, each cited to its source HG signal across the seed set)
- Look-alike table (up to {{cohort_size}} rows: company, domain, industry, revenue band, matching signals, score)
- Confidence note: how thin the signature is, which patterns dominated
## Quality Checklist
- Every match cites the patterns that placed it on the list, not just a similarity number
- Excluded domains from {{exclude_domains}} are visibly absent
- If the signature is sparse (fewer than 3 strong shared features), say so before listing matches
- No fabricated firmographic bands; use HG-returned values
