“Why is my KYC still pending?” If you run support for an online broker, you’ve answered that question thousands of times. It’s also the question an AI agent is most likely to get wrong. A general explanation of KYC sounds helpful, but it tells the customer nothing about their own application.

When we started working with FXTrading.com, the online trading platform of Australia’s Gleneagle Group, that question became our design brief. Together with FXTrading’s product and service teams, we helped connect Fin AI Agent to the broker’s internal systems and its Sumsub verification flow, and designed support flows around the questions clients actually ask.

The results, as reported by FXTrading.com: Fin sends its first reply within 3 seconds on average, resolves 84.27% of the conversations it’s involved in, and customer satisfaction stands at 91.1%. On the verification side, KYC now takes about 72 seconds, down from 49 minutes to 4 hours of manual review.

Here’s how we approached the project, and what we’d recommend to any regulated business deploying Fin.

Key takeaways:

  • Fin replies within 3 seconds on average and resolves 84.27% of the conversations it’s involved in.

  • 91.1% CSAT, with an average 7-second first human response when Fin hands a conversation to the team.

  • KYC in about 72 seconds with Sumsub, down from roughly 49 minutes to 4 hours of manual review (excluding resubmissions).

  • Core design principle: Fin gets customer context through FXTrading’s own internal API, not directly from Sumsub, so the client stays in control of data, procedures and handovers.

  • Fin features in use: Procedures for KYC status and upload issues, Workflows for handover, approved knowledge, Recommendations and Monitors.

[Embed video here] Figures reported by FXTrading.com; service metrics cover June 4, 2024 to October 28, 2026.


Project at a glance


Details

Client

FXTrading.com, part of the Gleneagle Group (founded in Sydney in 2010); regulated by ASIC in Australia, with eligible international clients served by a VFSC-licensed entity

Client scale

Operating since 2014; 100,000+ traders; 200+ financial instruments

Our role

Connecting Fin and Sumsub to FXTrading’s systems and processes; designing support flows and handovers

Stack

Fin (AI customer agent), Sumsub (identity verification), Salesforce (CRM and marketing), FXTrading internal systems and API

Fin capabilities

Procedures, Workflows, approved knowledge, Recommendations, Monitors

Results

First reply within 3s; 84.27% resolution; 91.1% CSAT; 7s human first response; KYC in ~72s


The client: a broker that builds its own products

FXTrading.com has served more than 100,000 traders since 2014, across more than 200 financial instruments. Unusually for a broker its size, it builds its own products: the trading app, Web Trader, social trading, funds management, the client portal and the partner portal were all developed by its own engineers, and it is now developing FXT AI around trading behaviour, position risk and market analysis.

That mattered for our project. Because FXTrading owns its systems and APIs, we could connect Fin to the real source of customer context instead of bolting an AI agent onto the side of the business.


What we found: four gaps between the product and the service

We start every Fin deployment by mapping the customer journey before configuring anything. At FXTrading, that journey runs from discovery and registration through verification, trading and support. Reviewing it together, we identified four areas that needed attention:

“Sumsub supports verification. Fin helps answer customer questions. DKM helps us connect those services to our systems and processes so the client can move through the experience with a clearer next step.”

— Sam Choi, Chief Operating Officer, FXTrading.com


How we designed the Fin deployment

1. Give Fin the customer’s context, through the client’s own API.

Fin doesn’t query Sumsub directly. It retrieves customer context through FXTrading’s internal API, the same layer that connects the broker’s own systems. We favour this pattern for regulated businesses: there is one place to control what Fin can see, Fin’s answers reflect the account information the business already relies on, and the team can define the steps and handovers for each type of inquiry.

2. Write Procedures for the questions that matter most.

FXTrading has published Fin Procedures for KYC status questions and for technical problems such as camera and upload issues. Here is how Fin handles “Why is my KYC pending?”:

A customer with an upload problem gets different help from a customer whose application is simply waiting for review, and both leave knowing their next step.

3. Define the handover before you automate.

Inquiries involving identity documents, security or compliance follow a defined route to the right person. Documents are submitted only in the secure client portal, and verification decisions stay with FXTrading. When Fin hands over, the team’s average first response is 7 seconds, so the transfer doesn’t feel like starting again.

4. Make every conversation a source of improvement.

FXTrading uses Fin’s Recommendations to identify support content for review, and Monitors to flag selected conversations, including sensitive-data issues and repetitive loops. The team reviews what needs to change, so repeat questions become fixes rather than tickets.

How the stack fits together

System

Role

Fin

AI customer agent: answers, Procedures and routing to the team

FXTrading internal systems and API

Hold customer context; the layer Fin retrieves it from

Sumsub

Identity verification (KYC), with risk and AML screening where enabled

Salesforce

Customer relationships and marketing

DKM Ecosystem

Integration and support-flow design across the stack


Where Sumsub fits: KYC in about 72 seconds

Fin can only give a clear answer about verification if verification itself is clear. Before Sumsub, FXTrading’s team reviewed identity documents and proof of address by eye, which took roughly 49 minutes to 4 hours. Visual checks had limits when judging authenticity, and documents in unfamiliar languages could need outside translation.


Before Sumsub

With Sumsub

Time to complete KYC

~49 minutes to 4 hours

~72 seconds (excluding resubmissions)

Review method

Manual visual checks of ID and proof of address

Automated verification flow

Problem documents

Limits in judging authenticity; unfamiliar languages could need outside translation

Immediate feedback tells the customer what to fix

Supported workflows can be adjusted and tested before rollout as regional requirements change. Risk and anti-money laundering (AML) screening can form part of the configured process where enabled, and verification decisions remain subject to FXTrading’s policies and applicable requirements.


The results

Metric

Result

Fin’s average first reply

Within 3 seconds

Fin’s reported resolution rate

84.27% of the conversations Fin is involved in

Customer satisfaction (CSAT)

91.1%

Average first human response after handover

7 seconds

KYC completion time

~72 seconds (previously ~49 minutes to 4 hours)

Service figures cover June 4, 2024 to October 28, 2026, as reported by FXTrading.com. KYC timing excludes cases that require resubmission.

“DKM has helped us design these support flows around the questions clients ask, with access to a person when the case needs one.”

— Sam Choi, Chief Operating Officer, FXTrading.com


What’s next

We’re now exploring further Fin capabilities with FXTrading: Operator to help investigate problems and propose improvements, Trends to surface changing inquiry topics, and guided Fin support over the phone, always with a route to a person when a case needs one.

Our playbook: 5 lessons for deploying Fin in a regulated business

  1. Map the journey before you configure the agent. The right Procedures come from the real questions customers ask at each step.

  2. Put your own API between Fin and your vendors. It keeps data, permissions and procedures under your control.

  3. Answer with context, not explainers. “Why is my KYC pending?” has a different right answer for every customer.

  4. Design the handover as carefully as the automation. Decide which questions Fin answers, which go to people, and how compliance-sensitive inquiries are routed.

  5. Measure satisfaction, speed and resolution together, and treat repeat questions as product feedback.


Frequently asked questions

Q: Can Fin answer account-specific questions like “Why is my KYC pending?”

Yes, when it is connected to the business’s own systems. At FXTrading.com, Fin follows appropriate identity checks and then retrieves the customer’s context through FXTrading’s internal API, so the answer reflects that customer’s situation.

Q: Does Fin connect directly to Sumsub?

Not in this deployment. Fin retrieves context through FXTrading.com’s internal API rather than querying Sumsub directly, which keeps data access and procedures under FXTrading’s control.

Q: Is Fin suitable for regulated financial services?

It can be, with clear controls. At FXTrading.com, documents are submitted only in the secure client portal, verification decisions stay with the business, sensitive inquiries route to people, and Monitors flag sensitive-data issues. Each business should assess the setup against its own regulatory requirements.

Q: What results did FXTrading.com see with Fin?

91.1% CSAT, an average first reply from Fin within 3 seconds, a reported resolution rate of 84.27% of the conversations Fin is involved in, and a 7-second average first human response after handover (June 4, 2024 to October 28, 2026).

Q: Which systems can DKM connect Fin to?

Fin can be connected to verification platforms such as Sumsub and to CRM and help desk platforms including Salesforce, HubSpot, Zendesk and Intercom, typically through the business’s own API layer.

Q: What does DKM do in a Fin project?

We map the customer journey, connect Fin to your systems and verification provider, design Procedures, Workflows and handover rules with your team, set up monitoring, and support you after launch.


Deploy Fin with DKM Ecosystem

Every Fin deployment we run starts with the same goal we had at FXTrading: customers should always know where they stand and what to do next. We can help you with:

DKM Ecosystem is a global RegTech systems integrator and enterprise SaaS solution provider serving high-growth businesses in Financial Services and Fintech, Retail and E-Commerce, Internet and SaaS, and Gaming/iGaming, with 24/7 global KYC technical support after launch.

Planning a Fin deployment, or want more from the one you have? [Book a Fin consultation] and we’ll map your customer journey with you.

▶ Watch FXTrading.com tell the story: FinAI x Sumsub UseCase | FXTRADING.com: Connected Customer Experience

Figures in this article were provided by FXTrading.com for the periods stated and reflect its own configuration; results vary by business. Published with FXTrading.com’s permission. Product names are trademarks of their respective owners.




DKM Ecosystem is an Intercom APAC Distributor, APAC Gold Partner, and Professional Services Certified Partner, dedicated to helping enterprises and partners adopt, implement, and scale Intercom across the Asia-Pacific market.

DKM Ecosystem delivers Intercom consulting, implementation, integration, and professional services—enabling businesses to build, deploy, and continuously optimize AI-powered customer service systems. Its service capabilities span Fin, AI Agents, workflow automation, omnichannel customer service, system integration, localization, training, and technical enablement—providing end-to-end Intercom support from initial solution design and product deployment through ongoing operational optimization.

🌏 Professional services and localization expertise across APAC and key global markets: Including Mainland China, Singapore, Japan, South Korea, Hong Kong (China), Taiwan (China), Malaysia, the Philippines, Thailand, India, Indonesia, Vietnam, Cambodia, the UAE, and Australia

📡 Omnichannel customer service solutions: Supporting customer interactions across Chat, Email, Voice, WhatsApp, Instagram, Facebook Messenger, SMS, Slack, LINE, WeChat, Lark, DingTalk, and more

🔗 Intercom, Fin, and AI Agent integrations: Seamlessly connect with existing customer service platforms, CRMs, APIs, and core business systems

🤖 AI-powered customer service capabilities: Combine Fin, AI Agents, and workflow automation to elevate service efficiency and automation

🔑 Solutions for 35+ industries: Tailored AI, automation, and real-time analytics capabilities to meet diverse industry-specific customer service needs

📈 Training, technical enablement, and continuous optimization: Help businesses fully realize Intercom’s platform value—and continuously improve customer experience and service efficiency

If you’d like to learn more about Intercom, Fin, AI Agents, and AI-powered customer service solutions, or explore a customized Intercom implementation and integration plan for your business, start your free Intercom trial today, or contact DKM Ecosystem to discuss the right solution for your business.

👉 Learn more about DKM Ecosystem on the official Intercom Partner page:

https://www.intercom.com/solution-partner-program


Recommended Reading

DKM Ecosystem as Intercom APAC Distributor (Excluding ANZ), Driving AI Customer Service Adoption Across Asia Pacific

DKM Ecosystem Deepens Its Expertise in Intercom, Leveraging APAC Gold Partner Status to Scale AI-Powered Customer Service Across Asia Pacific

Intercom Practical Guide (1): Why Do WhatsApp Messages Fail When Sent via Intercom?

Intercom Practical Guide (2): Does “User” on Mobile Really Mean a Logged-in User?

Intercom Practical Guide (3): How to Integrate Service Hub with the Intercom Ticketing System