Competitive map · enterprise AI & analytics services
Five firms, one market - and where each one actually stands.
An even-handed read of MathCo, Fractal, Tiger, Tredence and LatentView in enterprise AI & analytics services - each firm assessed from its own chair, on public evidence.
The five
The verdict
Independent read on public figures (cited in each firm's tab and the methodology). The five-year war-game below is an illustrative, fully adjustable model - not a forecast.
The market
Enterprise AI & analytics services is consolidating around a handful of specialists. These five - from a focused challenger to scaled, public leaders - compete for the same Global-2000 wallet: data engineering, advanced analytics, GenAI, MLOps and decision science.
Both rates are cited anchors in the war-game below. Multi-sourcing - several of these vendors inside one client at once - is the documented norm.
How to read this
Each firm has its own tab, written from that firm's chair - its position, where it can grow, and what it should fear. This page is the shared field: who leads each slice of the wallet, and how the five compare on scale and channel. MathCo is one of the five, assessed on the same terms as the rest.
Who leads each slice
The wallet, and who holds each part of it.
A typical enterprise analytics+AI wallet splits across service lines. No single firm leads them all - each has a home slice and cedes others, which is what makes the field contestable.
Badge legend · how every number is grounded
Scale & channel
Size on one axis, hyperscaler channel on the other.
Revenue scale and hyperscaler co-sell leverage separate the five more than service mix does - and the two don't always move together.
Fractal and Tiger anchor the high-revenue end; Tredence punches above its size on channel; LatentView and MathCo are the smaller, more focused players. Scale and co-sell don't always move together. (MathCo's revenue dot is an India-entity filing that likely understates its global billing, so read it as a floor, not a precise rank.)
X: best-available public revenue, which mixes fiscal years and entity scopes - MathCo's is an India-entity filing that likely understates its global billing; Tiger's and Tredence's globals are CEO-stated and unaudited - so read scale as directional, not precise. Y: a relative read of hyperscaler co-sell & platform leverage from public partner status (Databricks/Snowflake/Microsoft tiers) and analyst recognition, not a published metric.
The next five years
If the market keeps growing, who captures it?
An illustrative, fully adjustable war-game: set how hard each firm pushes and how fast the market grows, then watch the five-year trajectories - and how much of an expanding market the five capture versus everyone else.
Two lenses, never added.
The ~$18B services slice sits within / alongside the ~$85B data-&-analytics market. Different research firms, so they overlap rather than strictly nest - switching the benchmark only re-frames the five's slice; their revenue is identical in both views.
How hard does each firm push? Starting points reflect each firm's own stated ambition - drag to your own view.
The two rates measure different market definitions (broad AI services vs the narrow consulting slice this projection's $18B TAM comes from) - not a disagreement about one number.
What moves this conclusion - and what flips it
One-way sensitivity of MathCo's 2030 revenue: each driver swung across its range (cited range for market growth, full range for postures and constants), others held at your settings. Badges say how each driver is grounded.
Flip point · MathCo holds share
MathCo loses share ($172.2M vs $231.5M needed). Flips only at aggressiveness ≥ 54 (now 35) - roughly well above its stated posture.
Robustness · both cited growth anchors
The share verdict is the same at Gartner's 21.4% and R&M's 31.6%: MathCo loses share under both definitions - the conclusion doesn't depend on which research house you buy.
Ambition check · "$1B by 2030"
Tiger: $1.5B - goal met here
Tredence: $1.5B - goal met here
Both goals are the firms' own public statements - this model lets you test them against the cited market growth.
Your current sliders (postures, market growth, model constants) are sent to the server, re-clamped and recomputed there, then critiqued. Normally that critique is written by a live model, restricted to the engine's own numbers. If no model is configured, or it cannot answer usably, you get a deterministic scripted review instead - and the panel says which one you are reading, plus why.
Ease Tredence off the gas (aggressiveness → 0) and its 2030 revenue drops from $1.5B to $446.2M - while its rivals capture +$694.8M by 2030 that it cedes. Dial any firm back and the others - and the rest of the market - pounce.
At these settings the five capture ~6.9% of all new growth in the services market they sell into, through 2030; the rest goes to the Big-4, hyperscalers and in-house teams.
Mixed fiscal bases: MathCo FY25 · Fractal/Tredence/LatentView FY26 · Tiger CY2024.
What's real vs modelled
Company facts - revenue, profit, headcount, funding, clients, analyst standing - from filings, results releases and cited press (15+ sources below). Tiger/Tredence globals are CEO-stated, unaudited, and flagged as such.
Every dollar of opportunity, risk or projection is the output of visible, badged assumptions you can change - including the models' own mechanism constants. Sensitivity panels report which assumptions carry each conclusion, and its flip points.
The red-team critique is a live LLM restricted to the engine's computed numbers when a model key is configured - and a deterministic scripted review when not. The result is labeled either way; nothing scripted is ever presented as AI.
Methodology & sources