Leasing & Financial Services Automation

    Multi‑Funder Quotation Automation for a UK Leasing Company — From 16 Hours to Under 20 Minutes

    Client

    UK Leasing Company

    Process

    Multi‑funder Quotation

    Industry

    Automotive Leasing

    Outcome

    16 hours → under 20 mins


    The Challenge

    The client operates in a competitive UK vehicle leasing market where the speed and accuracy of funder quotes directly affect conversion rates. Customers expect rapid responses, and delays in the quotation cycle — even of a few hours — create meaningful commercial risk as customers seek alternatives.

    To generate competitive quotes, operators were required to manually interact with each funder portal in sequence:

    • Log into each funder portal individually using separate credentials
    • Navigate each portal's unique interface to submit vehicle specifications and pricing parameters
    • Wait for each funder to return a quote — timing varied by provider and time of day
    • Manually record and re-enter all quote results into the client's internal system for comparison

    The process was time-intensive, repetitive, and prone to human error at every transcription step. An earlier attempt at partial automation had reduced some manual work, but the end-to-end quotation cycle still took more than 16 hours — a fundamental commercial constraint.

    With the market moving to same-day or near-instant quotation cycles, a 16-hour process was not competitive. The business needed a step-change in speed, not incremental improvement.

    Why Funder Portal Automation Was Difficult

    This was not a simple integration task — it was one of the most technically challenging automation scenarios in UK vehicle leasing automation.

    The quotation process involved more than 10 external funder portals, each with distinct characteristics:

    • Completely different web interfaces, navigation flows, and form structures — no two portals worked alike
    • Individual authentication and session management requirements — including portals with multi-factor authentication
    • Varying response times — some portals returned quotes in seconds, others required minutes of processing
    • High-volume quote requests under commercial time pressure — the system had to be fast enough to deliver competitive advantage, not just automate the existing slow process
    • Portal interfaces that change without notice when funders update their systems — a maintenance challenge for any automation built on top of them

    Traditional sequential automation could not deliver the required speed. Even with perfect reliability, processing 10+ portals one after another produces a runtime too long to be commercially useful. The solution required a fundamentally different execution architecture.

    Any solution had to operate reliably in production, handle portal changes without catastrophic failure, and complete the full quotation cycle in a timeframe that changed the commercial equation.


    The RINKT Implementation — Parallel Browser Execution at Scale

    RINKT implemented a multi-funder quotation system that automates quote retrieval and consolidation across all funding providers simultaneously. The key architectural decision — and the reason the system achieved the performance target — was parallel browser execution: multiple independent browser instances operating concurrently, each handling a separate funder portal, rather than processing portals sequentially.

    What the system does — step by step

    • Receives quotation parameters from the client's internal system — vehicle specifications, contract terms, and pricing inputs
    • Launches separate browser instances for each funder portal simultaneously, managing credentials and session authentication independently per portal
    • Submits the required parameters into each portal's unique interface, handling the distinct navigation flow and form structure of each provider
    • Monitors each portal concurrently, retrieving results as they become available rather than waiting for all portals to complete before proceeding
    • Validates each retrieved quote for completeness and internal consistency before accepting it into the consolidated result set
    • Normalizes quote data into a consistent format and updates the client's internal system automatically with the full comparison set
    • Logs exceptions — portals that failed, timed out, or returned unexpected results — for review without holding up the rest of the quotation run

    The result: a complete multi-funder quotation cycle — across all 10+ providers — completed in under 20 minutes. Previously, the same process took over 16 hours.

    Key Technical Decisions

    Several technical decisions were essential to achieving this performance reliably in production:

    Controlled concurrency rather than maximum parallelism

    Running too many browser instances simultaneously risks triggering funder portal rate limits, session lockouts, or anti-bot detection. RINKT implemented controlled concurrency — a defined maximum number of simultaneous sessions per portal, with intelligent queuing when that limit is reached. This balances execution speed against the risk of provider-side blocking. The concurrency parameters were calibrated through testing against each funder's actual behavior before go-live.

    Independent failure domains per portal

    In a sequential system, a single portal failure could halt the entire quotation run. In the parallel architecture, each portal operates in its own isolated execution context: a failure or timeout in one browser instance does not affect the others. The quotation run completes with the results available from portals that succeeded, while flagging exceptions for review. This design means the system delivers commercial value even on days when one or two portals have issues.

    Portal-specific automation modules with shared orchestration

    Each funder portal has its own dedicated automation module — a self-contained set of instructions for navigating that portal's specific interface. These modules share a common orchestration layer that manages execution, result collection, and error handling. When a portal interface changes, only the affected module needs to be updated — the orchestration layer and all other portal modules remain stable. This modular architecture significantly reduces the maintenance burden of funder portal automation.

    Result normalization for downstream comparison

    Different funders present quote results in different formats, with different field names, rate structures, and assumptions. A raw quote dump from 10 portals would be difficult to compare without manual interpretation. RINKT built a normalization layer that converts each portal's raw output into a consistent internal format, enabling direct comparison without manual translation. This was essential to realizing the full commercial value of the automation — speed is only useful if the results are immediately actionable.

    Implementation Focus

    The implementation was designed with production constraints as the primary driver — not feature completeness:

    • Controlled concurrency calibrated to avoid provider throttling or account lockouts
    • Full visibility into execution status — operations and IT teams can see which portals are running, completed, or in exception at any point during a run
    • Modular portal modules enabling rapid updates when portal interfaces change, without disrupting the wider system
    • The ability to run quotations multiple times per day — the system is designed to complete a full run quickly enough that it can be executed at regular intervals throughout the trading day

    Business Impact Delivered

    With the multi-funder quotation automation live in production, the client achieved transformative results:

    • Quotation cycle reduced from 16 hours to under 20 minutes — a 98% reduction in elapsed time for the full multi-funder comparison
    • Over 80% reduction in manual effort associated with the quotation process — operators no longer spend hours working through portal queues
    • Ability to run quotations multiple times per day — the client can now refresh quotes mid-day when funder rates change, previously impossible with a 16-hour cycle
    • Reduced data entry errors and rework — automated extraction and normalization eliminates manual transcription from portal screens into internal systems
    • Improved commercial responsiveness — customer queries can be answered with fresh funder data rather than quotes from the previous day's run
    • Full transparency for operations and IT teams — every run produces a complete execution log with timing, results, and exceptions per portal

    Quoting shifted from a bottleneck that constrained the business to a scalable operational capability that supports commercial growth.

    Strategic Value

    Beyond the immediate efficiency gains, the multi-funder quotation system delivered lasting strategic advantages for the leasing business:

    • Improved competitive positioning — the client can now match same-day quotation cycles that larger competitors offer, without the overhead of large portal teams
    • Better pricing decisions — access to fresh, comparable funder data multiple times daily enables more responsive pricing strategies
    • A reusable automation pattern applicable to new funder relationships — adding a new funder portal requires building a single new portal module without redesigning the system
    • The implementation has demonstrated the viability of parallel portal automation across adjacent leasing workflows

    The implementation laid the foundation for further automation across the leasing workflow — the parallel execution architecture and portal module pattern applies broadly beyond quotation.


    Frequently Asked Questions: Multi-Funder Quotation Automation

    How many funder portals can this system handle?

    The system in this implementation covers more than 10 funder portals simultaneously. The parallel architecture does not have a fixed upper limit — additional portals are added by building a new portal-specific module without changing the orchestration layer or existing portal modules. In practice, the limiting factor is the number of funder relationships the leasing business has, not the technical capacity of the automation. We have also implemented similar systems for clients with fewer funders where the quotation cycle time, rather than the number of portals, was the primary commercial constraint.

    What happens if a funder portal changes its interface?

    Portal interface changes are one of the key maintenance challenges in funder portal automation. RINKT's modular architecture minimizes the impact: each portal's automation logic is isolated in its own module, so a change to one portal only requires updating that module — it does not affect the others. Monitoring detects portal failures immediately, so interface changes are identified quickly rather than discovered through downstream errors. Typical resolution time for a portal interface change is measured in hours, not days. Where portals offer APIs — even undocumented ones — RINKT will use these in preference to UI automation to reduce sensitivity to interface changes.

    Does this automation replace manual quoting entirely?

    For standard quotation parameters, the automation handles the entire retrieval and comparison process without human involvement. For unusual vehicle specifications, portals that require human judgment calls, or cases where a funder portal produces unexpected results, exceptions are routed to operators with full context. In practice, manual involvement is reserved for genuinely complex or unusual cases — typically a small fraction of total quotation volume. The automation does not replace the commercial judgment that goes into what to do with quotes; it eliminates the mechanical work of retrieving and collating them.

    How long does it take to implement a multi-funder quotation automation?

    Implementation timelines depend on the number of funder portals, the complexity of each portal's interface, and the scope of internal system integration required. For a system covering 10–15 funder portals with standard internal system integration, a production-ready implementation typically takes 10–16 weeks from process qualification to go-live. This includes a testing phase where the automation runs against live portals in parallel with the existing manual process, so results can be validated before the manual process is wound down. RINKT provides a detailed timeline estimate as part of the implementation planning engagement.


    Why This Worked

    This implementation succeeded where simpler approaches had failed because:

    • The process was fully qualified before automation — RINKT understood every portal's behavior, the full range of quotation parameters, and the internal system requirements before architecture decisions were made
    • Parallel execution was designed deliberately as a first-class architectural requirement — not retrofitted after discovering that sequential processing was too slow
    • Control and reliability were prioritized over raw speed — the concurrency model was tuned to stay within provider tolerance rather than optimized purely for minimum runtime
    • Automation was built for production, not demonstration — tested under real portal load before the manual process was decommissioned

    This is the difference between automation that looks good in a demo and automation that delivers competitive advantage in production.


    Related Implementation Pattern

    Multi‑Funder Quote Normalization

    The implementation pattern behind this case — parallel portal automation with result normalization — is applicable across leasing, asset finance, and other financial services environments where multi-provider comparison is a core operational requirement.

    → View Implementation Pattern

    Multi-Funder Quoting Taking Too Long?

    If your leasing or asset finance operation relies on high-volume quotation across multiple external funders and the manual process is constraining your commercial responsiveness, RINKT's parallel portal automation approach may apply. We implement production-grade multi-funder quotation systems that deliver real-time comparison at scale.

    Get Your Implementation Plan