How Brokers Grow: Webinar Takeaways and Audience Q&A

On October 15, Devexperts hosted a live panel on how brokers grow when acquisition costs keep rising and clients have more venues to choose from. The panel brought together four people who see the problem from different sides: Jon Light, Senior Director of Product Management at Devexperts, who moderated; Ed Hancock, Director of Operations at Pelican; Rauan Khassan, Chief Growth Officer at TradingView; and David Kimberley, founder of TradeInformer. 

Over the hour, they compared what separates growing brokers from stalling ones, how prop trading works as an acquisition channel, what keeps clients active after the first deposit, how to market on a limited budget, and where AI stands today. This article summarizes the discussion and answers six audience questions the panel had no time to reach. The full session is available on Devexperts’ YouTube channel.

What growing brokers do differently 

David opened with a split he has watched form over the past decade. A smaller group of well-funded firms has moved toward regulated markets, where acquisition costs and the regulatory burden both run higher, and has built multi-asset products that put every instrument in one app. For most of them, he said, going multi-asset now feels like a condition of survival. The rest of the sector has stayed with the core FX/CFD product and grows mainly through partners. 

Consistent execution across the product

Within either group, brand and timing explain part of the difference. David pointed to the commission-free trading apps that reached the UK and US markets first. They built their positions on that head start.

The pattern he returned to most often, though, was consistent execution. He described one of the largest CFD brokers as an example. Its sign-up flow, client portal, partner portal, trading conditions, and in-house platform all work well. Small firms struggle to match that combination. One product manager often juggles twenty projects, so no single feature gets the attention it needs.

Focus before adding products

Ed warned against chasing the largest brokers feature by feature. Any new product, copy trading included, needs someone inside the broker who owns it and markets it alongside the brand. Ed noted that Pelican has “a CTO who says no to lots of things.” He recommended the same discipline to brokers. If partners and IBs drive your business, strengthen that core before adding more.

Onboarding and technology budgets

Rauan raised client onboarding. He holds accounts with many of TradingView’s integrated brokers and sees wide gaps in account opening times. Some brokers take minutes, and others take days, and the promised callback sometimes arrives after the weekend. On TradingView, brokers with the same contract terms get different results. In some cases, the cause sits in the onboarding flow on the broker’s own site.

He then made the case for technology budgets. Despite the attention AI attracts, he sees little budget moving from sales and marketing into technology. Sales results are easy to count. The effect of technology shows up in how long clients keep trading with you. A broker’s connection has to hold during the next NFP release. A failure while client money is at stake ends the relationship.

Feature depth and client quality

Rauan also pointed to gaps in basic functionality. Some brokers integrated with TradingView still do not support bracket orders. Rauan, an active trader himself, called that a showstopper. Once the basics work, feature depth sets a broker apart. In his experience, clients who rely on technical features trade more actively and fund larger accounts. Those are the clients sales and marketing teams want to reach.

Jon framed growth as a question of client quality as well as client count. The successful brokers Devexperts works with offer multiple asset classes and expand into regions they understand. They build relationships with the local IB market there and match their technology to the client segments they target.

Prop trading as an acquisition channel 

Over the past few years, prop firms have multiplied, and some brokers now run prop brands to acquire clients. David said several brokers treat prop purely as a funnel. Because prop challenges generally sit outside brokerage regulation, firms can advertise them more widely on search and social platforms. The difficulty comes afterward, when the broker tries to convert challenge participants into brokerage clients. In David’s view, no one has done this well yet, a point he also makes in a recent TradeInformer Prop Weekly column. 

He finds the gap odd, since the products are close and the client’s goal is similar. He compared it to crypto, where traders accept perpetual futures but reject the same exposure when it carries a CFD label. Broker trading desks struggle with prop for a related reason. Prop behaves like an iGaming product, so experience in netting positions and hedging residual exposure transfers poorly.

Ed described prop as a test bed. Pelican sees brokers bring successful prop traders onto the brokerage side, where they join the copy trading network as strategy providers and earn performance fees. He admitted that deciding how to split resources between prop and brokerage is hard for any organization. 

How prop marketing has changed

Rauan said prop marketing has improved since the first wave of growth about three years ago. Firms have dropped vague claims about a multi-million fund traded on a demo server. They now describe a challenge as what it is: a competition with stated criteria and a payout for those who meet them.

TradingView has no prop section of its own. It supports prop activity through its integrated brokerages and treats the broker’s license as reputational collateral. From the trader’s side, Rauan added, €500 buys several challenge attempts. The same amount in a live account leaves little room to build anything consistent. 

What keeps clients active after the first deposit 

Once a client has deposited, the work shifts to keeping them trading. Ed said many new clients stall at the first trade. Copying an experienced trader removes that barrier and teaches by example. Pelican’s network carries more than 9,000 live strategies from over 60 brokers, and followers often become loyal to top strategy providers. When a strategy hits a losing stretch, Pelican sees many followers deposit again, either to return to the same strategy or to move to one that trades different markets. 

Jon asked whether brokers can market themselves through the traders their clients copy. Ed explained that the network’s content is shared across connected brokers, so a small broker can offer a large pool of strategies from day one. Each broker still needs the product tailored to its clients, and Pelican monitors the shared content so the statistics hold up and short-lived accounts do not crowd the rankings. Pelican’s revenue depends on its brokers doing well, Ed said, so the team stays involved after launch. Brokers licensing DXtrade can offer Pelican’s copy trading from within the platform. 

Marketing on a limited budget 

For brokers without large budgets, David said, the default route is IB or CPA-based growth, and most small firms already take it. The harder step comes later: moving from partner-led growth into brand and organic acquisition. Firms raised on partner economics expect every dollar spent to show up quickly as deposits, so brand work, where results are hard to trace and some spend is always wasted, feels wrong to them. 

Until then, David’s advice is to make the product work: dashboards that serve IBs and clients well, and onboarding that is quick to complete. He described a broker that lost a marketing partnership because it could not track attribution, which left the partner no way to measure results. 

Ed added that the IB model relies on personal contact, with someone guiding each client through the process. Brand-led acquisition works at arm’s length, so the process has to be straightforward and the experience right after the first deposit has to go well. A client whose money seems to vanish straight away, with no explanation, won’t come back or recommend you to anyone. 

Rauan then covered how brokers use TradingView. Most treat it as a performance marketing tool, which he called correct but incomplete. For mid- and lower-tier brokers, he recommended subscription coupons. With more than 90% of TradingView users on free plans, he said, a paid subscription works as a commodity with a strong call to action. In his estimate, manually distributed coupons get activated only 20% to 25% of the time, because emails land in spam or clients forget. TradingView automated the process through an API endpoint: the broker defines conditions in its CRM, such as trading volume or trades in particular symbols, and when a client meets them, the subscription activates for the next 30 days.

Standing out as a brokerage

For brokers that want to stand out, Rauan described a project TradingView is close to completing: giving clients of integrated brokers access to its MCP server, which uses the Model Context Protocol to connect TradingView accounts to AI assistants. TradingView introduced the server in beta in September. 

Brand campaigns on TradingView, whose audience Rauan put at about 100 million users, are a different matter. They suit top-tier brokers willing to pay for visibility they cannot track. Rauan cited an e-commerce rule of thumb that a person needs about seven touchpoints before acting, and described campaigns that place a broker’s name across a client’s chart and screen so it is the first option in view when the client decides to trade. 

A broker that already licenses DXtrade can expect a TradingView integration to take about two months, unless the project involves custom work. And to improve organic visibility on TradingView, brokers should start replying to client reviews. 

Where AI stands for brokers 

Jon left AI off the agenda on purpose, but audience questions brought it back. David compared the current moment to mobile trading in the mid-2000s: a new technology, parts of which still feel like gimmicks. Its present value lies in removing manual steps, much as an AI assistant spares you a spreadsheet lookup. He does not yet see it reaching the point where a client builds a personal frontend and hands trading over to it entirely. 

Jon said MCP connections suit some tasks and not others. A large language model is slow and cannot take in large volumes of live data, which rules it out for fast execution. It is useful for analyzing historical data and for turning that analysis into strategies and trading bots. Prospects now ask Devexperts whether they can control the platform through AI, and DXtrade answers with a trading control API that brokers have connected to AI assistants through MCP servers. 

Jon also expects clients to bring their own interfaces to trading. As AI and MCP services sit on top of broker APIs, clients will pick a platform and a broker separately, instead of choosing a broker for its platform. 

On retention, Jon took an audience question about the early signs that a client is about to go inactive. DXtrade’s AI retention and engagement tool analyzes client engagement and predicts churn, flagging clients likely to withdraw funds and leave while there is still time to act. Retention teams receive these predictions as regular reports or through an API that feeds the broker’s CRM. 

One piece of advice from each panelist 

Jon closed by asking each panelist for one action brokers can take next. 

Audience questions we didn’t reach 

Attendees sent more questions than the hour allowed. Below, we answer six of them. 

What are the advantages of a SaaS platform model, and does it leave enough room to grow and change focus? 

A SaaS (software as a service) trading platform is licensed as a subscription: the vendor runs the hosting and the updates, and the broker pays a recurring fee. The main advantage is speed to market. A new broker can go live without building infrastructure or hiring an engineering team, and can spend its early budget on acquiring clients. Our guide to starting a forex brokerage walks through the launch roadmap. 

Room to grow depends on two factors: how open the platform is and whether the vendor offers a path beyond SaaS. With open APIs, you can add a CRM, a copy trading service, an AI tool, or a second frontend as needs change, without replacing the core. With an enterprise option from the same vendor, you can move to a customized deployment once your volumes justify it. 

DXtrade follows this model. It is a turnkey white-label platform with the option of partial or full customization, and its open integration framework lets brokers add third-party services as they grow. As Jon noted during the webinar, Devexperts can start a broker on a SaaS configuration and move it to enterprise as it scales. For more on choosing between one supplier and several, see our article on what AI changes for broker technology. 

Fintech provision has moved from walled gardens to modular infrastructure. What drove the shift, and where is it heading as providers expand beyond their original focus? 

Two forces drove the shift. Brokers learned the cost of vendor lock-in: with an all-in-one platform, you get the features your supplier builds on its schedule, and switching grows expensive once the business depends on the bundle. Meanwhile, specialist vendors built integrations with one another and now expect to run side by side, so a modular setup is easier to assemble than it was ten years ago. 

The webinar panel showed this model at work. DXtrade connects to TradingView for trading from charts and to Pelican for copy trading, while each company develops its own layer. 

Expansion beyond a provider’s original focus is already under way: copy trading providers offer broker-branded apps, and charting platforms connect to AI assistants. As the overlap grows, judge vendors by how openly their APIs let the pieces work together. License the regulated core from specialists (the trading platform, CRM, liquidity, risk, and payments) and add components where your business needs them. Your starting model depends on your stage: a new broker with no engineering team often begins with a single supplier, then adds components as volumes and client types multiply. Our e-book on balancing in-house and outsourced development covers this decision in more depth. 

What technological hurdles do brokers commonly face during periods of rapid growth? 

Growth exposes the parts of a setup that worked at a smaller scale. Five hurdles come up most often. 

• Stability under load. Volume peaks around major releases such as NFP, and a platform that falters at those moments loses clients when their money is at stake.
• Onboarding throughput. Verification steps that run quickly at low volume can stretch to days when sign-ups multiply, and a client left waiting goes cold, as Rauan described on the panel.
• Payments. Payment connections are a constant source of day-to-day trouble in brokerage operations, and failed deposits at scale surface as support tickets and drop-offs.
• Dealing configuration. As client segments multiply, dealers need to change trading conditions for groups of clients quickly. In DXtrade, dealers assign profiles for commissions, limits, margins, and spreads to an account or group, and changes take effect without a server restart. DXtrade’s risk management tools add exposure limits and tiered margin settings on top.
• Measurement. A broker that cannot track attribution cannot work with partners paid on results, as David described on the panel.

How can brokers increase account funding rates and promote responsible trading to raise LTV? 

The funding rate is the share of registered accounts that make a first deposit. Client lifetime value (LTV) is the revenue a client generates over the whole relationship. Both improve when you remove friction and when clients keep trading long enough to build a relationship. 

To raise the funding rate, measure where applicants drop out between registration and first deposit, and fix the slowest step first. Help delivered inside the platform at the point of friction keeps sign-ups moving. Devexperts’ behavioral analytics work identifies repeated deposit failures as a moment when a client is receptive to help. 

For responsible trading, a platform that sets a stop-loss on new positions by default, which the client can keep or adjust, standardizes risk management and teaches it at the same time. Stating risk in currency terms also helps: telling a client how much they stand to lose before the stop triggers puts the risk in terms they grasp at once. Our e-book on UX and risk management covers interface patterns such as validation and alerts. Copy trading gives new clients a structured first trade, as Ed described on the panel. 

Responsible trading and LTV point in the same direction, because LTV accrues only while a client keeps a funded account. 

Do you see AI becoming part of traders’ daily workflow, and is this a generational trend? 

Adoption is already high among younger investors. The UK’s FCA said in October that four in five less-experienced investors aged 18 to 40 have used AI for investing. The FCA data covers younger investors, but the tasks AI handles well today, research and historical analysis, serve traders of any age, so we expect use to spread across age groups. 

For now, AI sits mostly in the research layer of trading. TradingView’s MCP server, for example, offers market data and research tools but no access to broker positions or order placement. For brokers, the task is to make the platform reachable through AI on the client’s terms while every order still passes through existing risk controls. 

FCA findings also show a gap: only 40% of consumers understand the limits on recourse after acting on advice from general-purpose AI tools. Brokers that explain what their AI features can and cannot do give clients a reason to trust them. 

Is AI giving brokers meaningful improvements in retention, and what would an ideal implementation look like? 

AI changes when a broker contacts a client and which clients it contacts. Models trained on trading data and platform activity can segment clients, track engagement, predict churn, and assess the riskiness of trading styles. Retention work can then start before a client leaves. 

An ideal implementation has four parts: 

• Grounding in the broker’s own data, so answers and alerts reflect your configuration and your clients.
• In-platform delivery timed to client activity. A message about deposits works best when the client is viewing balances, and midweek sessions suit educational content.
• A human in the loop for any decision with financial or legal weight. With 86% of UK financial services firms now using AI, the FCA has urged regulated firms to maintain human oversight of automated systems.
• Measurement against a control group, so your retention team knows which interventions change client behavior.

DXtrade’s AI retention and engagement tool combines a broker-facing layer, which segments clients and predicts churn, with a client-facing AI assistant inside DXtrade that delivers timed messages and gathers preference data from client questions.