Plug in your monthly investigation volume. Without Tangos, every additional case adds to the hiring curve. With Tangos, your team handles more — without growing at the same rate.
—
Click any assumption to expand the logic and tune.
Manual team scales with case volume. Each case type takes a different amount of analyst time:
EDD · 12 hrs · AML L2/L3 · 8 hrs · Sanctions L2/L3 · 5 hrs
At 130 productive hours per FTE per month (after meetings, training, PTO, alert triage, ad-hoc work), total monthly hours ÷ capacity gives total FTEs needed. Team splits 35% senior / 65% junior, which matches what most US institutions actually run.
Floor staffing: 1 senior + 1 junior on both sides at very low volumes (2 FTEs minimum). Even a small investigation function needs at least one senior reviewer and one analyst.
Manual operations scale aggressively in the first 150 cases / month — the manual floor adds 1 FTE per ~25-case threshold, because a small team can't actually absorb that growth without hiring. At 25 cases the floor is 3 FTEs; at 75, 5 FTEs; at 125, 7 FTEs. Above 150, natural team-size calc takes over.
Tangos stays at the 2-FTE floor through this entire range — the platform absorbs the ramp. Differentiation is visible from the first investigation.
Percentage applied to direct labor cost to capture indirect costs: benefits, office space, IT support, training, and management overhead. 40% is a common assumption for US financial institutions. Applied to all analyst labor on both sides.
Each Tangos-equipped analyst handles ~4× the cases of a manual analyst. Tangos automates the routine investigation steps — data gathering, screening, narrative drafting, packaging — so analysts spend their time on judgment and complex cases instead of toil. Your Tangos team is therefore ~1/4 the size of an equivalent manual team for the same volume.
Conservative: 2.5–3×. Aggressive: 5–6×. Your actual multiplier depends on case mix, complexity, integration depth, and adoption maturity.
The hiring curve story: manual operations need 1 more FTE per ~21 additional cases / month. Tangos at 4× productivity needs 1 more FTE per ~84 cases / month. As volume grows, the gap widens — that's where the savings come from.
Subscription scales with case volume. Four tiers cover small, mid-market, growth, and enterprise banks:
Subscription auto-tiers based on your volume. You can still override below for a custom price — but moving the volume slider will re-tier.
An autonomous financial crime investigation platform takes engineering, domain expertise, and a data foundation — recurring, year over year. Plug in your build assumptions and see the math against Tangos.
—
These figures reflect the recurring run-rate of operating a Tangos-equivalent platform internally. Building isn't a one-time cost — it's a recurring investment as models drift, data sources change, regulations evolve, and key team members leave. Year-one total is the floor, not the ceiling.
Building Tangos's investigation platform requires deep ML expertise across NLP (case narratives), graph reasoning (entity networks), and decisioning under regulatory constraint. The default 4 ML engineers + 1 ML scientist + 1 platform engineer is the minimum to ship a usable v1 in 12–18 months — and what's required to maintain it thereafter.
Salaries shown are base only. Overhead (benefits, equity, space, IT) is applied separately at 40% in assumption #02.
Percentage applied to direct labor cost to capture indirect costs: benefits, equity, office space, IT support, training, recruiting, and management overhead. 40% is a common assumption for US tech-and-financial-services compensation. Applied to all labor (engineering + domain expertise).
To meet examiner expectations, you need someone who has lived through SAR/EFR submission and OFAC enforcement actions — not just observed them. Ex-regulators with OFAC or Treasury AML leadership experience are a market of roughly a dozen people in the US. Compensation reflects the scarcity.
Tangos has this expertise embedded in the platform's reasoning, review workflows, and audit trails. Building from scratch requires sourcing it directly — this isn't a cost you can engineer around.
Investigation models require commercial data licensing (LexisNexis, Refinitiv World-Check, D&B, sanctions feeds), training-data curation across 15+ specialized domains (typologies, jurisdictions, beneficial ownership, etc.), and continuous validation & recalibration as regulations and adversary patterns shift.
All three are recurring. Models degrade within months without active maintenance — this is not a one-time investment.
Annual cost of the Tangos investigation platform. $40,000 / month represents the Enterprise tier (100+ investigations / month). Smaller tiers exist at $15K / $25K / $30K monthly for Starter, Small, and Growth-stage banks. Final pricing depends on case volume and is tailored during commercial discussions.