Every brokerage produces numbers by the millions — ticks, trades, deposits, clicks. Very few brokerages are actually run by numbers. The typical operating rhythm is a marketing report over here, a platform report over there, a payments spreadsheet somewhere else, and decisions made on the metric that shouts loudest that week — usually registrations, occasionally volume, and in bad months, whatever the dealing desk says.
The result is predictable: money spent acquiring clients who never fund, retention problems discovered a quarter after they started, and IB deals renewed on relationship rather than return. This guide lays out the KPI set that operators who run tight brokerages actually watch — organized as a funnel, with formulas, the questions each metric answers, and the classic ways each one lies to you.
First Principle: One Funnel, One Source of Truth
Brokerage KPIs only work as a connected chain: visitor → registration → verified account → first deposit → active trader → retained, profitable client. Each stage’s metric is the numerator of the previous stage’s denominator; break the chain into departmental silos and every team optimizes its own stage at the next stage’s expense. Marketing celebrates cheap registrations that never verify; sales celebrates conversions that deposit once and vanish.
The precondition is boring but absolute: every event in that chain — ad click, registration, KYC status, deposit, trade, withdrawal, support ticket — must land in one system tied to one client identity. That consolidated record is precisely what a brokerage CRM exists to be, a point we made at length in the complete guide to forex back office and CRM platforms. If your KPI meeting starts with reconciling three exports, fix that before adding a single new metric.
Acquisition Metrics: The Top of the Funnel
Registration conversion and cost
What: visitors → registrations, and cost per registration by channel.
Why: baseline funnel health and channel comparison — nothing more. Registrations are the classic vanity metric: cheap to inflate with broad targeting and worthless in isolation.
Registration → verified (KYC completion rate)
What: share of registrations that complete verification, and time-to-verify.
Why: the first quality gate. A big gap between registrations and verified accounts means either junk traffic or onboarding friction; time-to-verify predicts deposit probability strongly — momentum dies in queues. This is the stage where automated document processing and well-designed approval queues, as described in our piece on KYC and AML workflows in forex CRM, show up directly in revenue.
FTD rate and CPA (cost per first-time depositor)
What: verified accounts → first-time depositors (FTD); marketing spend ÷ FTDs = true CPA.
Why: this is the real price of a client, and the only sane basis for channel decisions. Channels should compete on CPA and the downstream quality of the FTDs they deliver — cheap FTDs that churn in three weeks are expensive.
The lie to avoid: averaging CPA across channels. The blended number hides the one channel quietly eating half the budget.
Time-to-FTD and average first deposit
What: days from registration to first deposit; median first-deposit size by channel and region.
Why: both are early-warning quality signals. Lengthening time-to-FTD usually points at funnel friction or weakening traffic intent; falling first-deposit size flags a channel drifting down-market long before LTV data can confirm it.
Monetization Metrics: What a Client Is Worth
Net deposits
What: deposits minus withdrawals, per period, per cohort, per client.
Why: the single most honest volume-adjacent metric a brokerage has. Gross deposits flatter; trading volume can be churned; net deposits measure real money entrusted to you. Watch it by monthly cohort: healthy brokerages show cohorts whose cumulative net deposits grow for many months, not spike-and-decay.
ARPU and revenue per lot
What: total revenue (spread, commission, swap, and — if applicable — B-side result) ÷ active clients; revenue ÷ lots traded.
Why: ARPU tracks monetization efficiency over time; revenue per lot exposes pricing reality across instruments and account types. Falling revenue per lot with stable volume means your mix is shifting to thin-margin flow — something to know before quarter-end, not after.
The nuance: if you run a hybrid book, separate flow-based revenue from B-side trading result in every report. Mixing them lets one good B-book month mask a decaying flow business — the exact confusion we warned about when comparing models in A-Book vs B-Book vs Hybrid.
LTV (lifetime value) and LTV:CPA
What: cumulative revenue per client over their lifetime, computed by cohort; the ratio of LTV to acquisition cost.
Why: the metric that decides how much you may spend on growth. Practical operators use bounded windows — 6- and 12-month cohort LTV — rather than projected lifetime curves, which flatter endlessly. A 12-month LTV:CPA comfortably above 3:1 is the conventional sign of an acquisition engine worth scaling; near 1:1 you are buying revenue, not building a business.
Retention and Activity Metrics: Where Brokerages Are Won
Active-trader ratio
What: clients who traded in the period ÷ funded clients; track weekly and monthly actives separately.
Why: your funded base is inventory; the active ratio is how much of it is working. A growing funded base with a sliding active ratio is a brokerage filling a leaking bucket — growth masking decay.
Churn and resurrection rates
What: share of previously active clients going inactive per month (define inactivity concretely — e.g., no trade in 30 days); share of dormant clients returning.
Why: churn compounds against you the way retention compounds for you, and it is far cheaper to prevent than to replace. The operational playbook — behavioral triggers, timed interventions, win-back sequences — is one we detailed in How to Reduce Trader Churn Using CRM Automation; the KPI’s job is to tell you whether that playbook is winning.
Redeposit rate
What: share of FTDs making a second deposit within 30/60/90 days.
Why: the sharpest single predictor of client quality and, in aggregate, of channel quality. First deposits are bought; second deposits are earned — by execution quality, payout speed, and service. When we analyzed what actually drives loyalty in What Forex Traders Actually Want From a Broker, the behaviors that correlated with redepositing were operational, not promotional — this metric is where that shows up.
Withdrawal SLA
What: median and 95th-percentile time from withdrawal request to funds sent.
Why: the client-visible proof of your operational health, the strongest single input to reviews and word of mouth, and a metric worth publishing internally with the same seriousness as revenue. Slow tails (the 95th percentile) do the reputation damage; medians hide them.
Partner Metrics: The Channel That Lies Most Convincingly
Cohort quality per IB and net partner ROI
What: for each partner — FTDs, net deposits, active ratio, churn, and revenue of referred clients, minus rebates paid.
Why: IB volume alone is the most gameable number in the industry. Partners should be ranked by the net revenue and retention of their cohorts, and rebate tiers should reward exactly that. If your partner reporting cannot produce a per-IB cohort P&L, the tracking layer is the problem — the architecture we described in Multi-Level IB Management Systems exists precisely to make this a report, not a research project.
Risk and Operations Metrics: The Guardrails
- Exposure and concentration — net exposure per instrument and per client cluster against limits; watched intraday by whoever owns the book.
- Chargeback ratio — disputes ÷ card transactions, tracked against card-scheme thresholds; the number that protects your payment rails.
- Payment approval rate — successful deposits ÷ attempts, by PSP, method, and region; every recovered percentage point is nearly free revenue, which is the entire argument of multi-PSP routing.
- Support metrics — first-response time, resolution time, and contact rate per 100 active clients (a rising contact rate is a product problem wearing a support costume).
- Uptime and execution quality — platform availability, rejection rates, slippage distribution; invisible when good, existential when not.
Making KPIs Operational: Cadence Beats Dashboards
A KPI program is a meeting discipline, not a wall of charts. The pattern that works:
- Daily (15 minutes, ops): deposits, withdrawals and SLA, approval rates, exposure, incidents. Detect breakage fast.
- Weekly (leadership): the funnel — registrations → verified → FTD by channel, CPA, active ratio, churn flags, partner outliers. Steer spend and effort.
- Monthly (strategic): cohort LTV, LTV:CPA by channel, net deposit cohort curves, retention trends, unit economics. Decide what to scale and what to kill.
Three rules keep the discipline honest. Every metric has an owner — a person, not a department. Every review compares against target and trend, not vibes. And every automated report comes from the system of record, not hand-built spreadsheets — the report-and-trigger machinery we outlined back in Retail Forex CRM Reports and Triggers exists so the numbers arrive without a human copying cells at midnight.
Frequently Asked Questions
What is the single most important KPI for a forex brokerage?
If forced to one: net deposits by cohort. It integrates acquisition quality, trust, retention, and payout behavior into one honest number that is hard to game. But its diagnostic power comes from the chain around it — a falling net-deposit trend sends you to redeposit rate, churn, and withdrawal SLA to find the cause.
Which metrics are vanity metrics?
Any number that grows without the business improving: raw registrations, gross trading volume, app downloads, and — the industry favorite — total registered accounts. Each has legitimate diagnostic use inside the funnel; none belongs in a headline.
How do KPIs differ for a prop firm?
The funnel reshapes — challenge purchases replace FTDs, pass rates and payout ratios replace net deposits — but the discipline is identical: one funnel, cohort economics, and metrics owned by named people. The prop-side economics have their own traps, which we covered in What Brokers Should Know About Prop Firm Profitability.
What benchmarks should I compare against?
Treat industry benchmarks as sanity checks, not targets — they vary wildly by region, channel mix, and account type. Your own trend lines and cohort-over-cohort improvement are the benchmarks that pay. The exception: card-scheme chargeback thresholds and regulator-facing numbers, which are external and absolute.
The Bottom Line
A brokerage that tracks the full chain — from cost per verified client to 12-month cohort LTV, from redeposit rate to withdrawal SLA — makes hundreds of small decisions per quarter slightly better than a brokerage running on instinct and platform exports. Compounded, that difference is the gap between the operators who scale and the ones who plateau wondering why. The metrics are not exotic. The discipline of unifying the data and reviewing it on cadence is the whole game.
Request a Consultation on Building a Brokerage KPI Framework
Get expert guidance on defining the metrics that matter most for your brokerage—from acquisition and onboarding to deposits, retention, partner performance, and long-term client value. We’ll help you structure a KPI framework that supports better operational decisions instead of isolated reporting.
Together, we’ll review your current reporting processes and outline a measurement strategy aligned with your business objectives.
