R&D Management 14 min read

How Huolala Enterprise Tackles Large-Client Customization with Standardized Architecture

The article examines Huolala Enterprise’s challenges in onboarding high‑value corporate clients, highlighting the high costs, revenue potential, and complex customization demands, and proposes a standardized, modular architecture—including domain modeling, micro‑kernel plugins, and configurable feature management—to achieve scalable, maintainable solutions.

Huolala Tech
Huolala Tech
Huolala Tech
How Huolala Enterprise Tackles Large-Client Customization with Standardized Architecture

Business Background

Huolala Enterprise version is a freight platform for corporate customers, offering same‑city delivery, customized city distribution, API support, cargo insurance, management functions, and monthly settlement accounts, serving over 1.5 million enterprises across retail, fresh food, furniture, e‑commerce, and other sectors.

Key Enterprise Clients

Large corporate clients bring significant long‑term revenue but involve high onboarding costs. The challenges are summarized in three aspects: high cost, high revenue, and demanding customization.

1. High Cost

Long Process : Typically 1–2 months, involving many roles and high communication overhead.

Many Demands : Diverse client requirements consume substantial product‑research resources; in H1 2023, large‑client demands accounted for 32 % of total freight demand.

2. High Revenue

80/20 Principle : A small number of large clients generate most of the revenue, with high per‑client contribution.

3. Challenges from Large‑Client Demands

High maintenance cost due to customized requests across business and industry differences.

Low iteration efficiency because core logic and differentiated code interfere, raising change cost.

Stability challenges: distributed custom implementations increase risk of system‑wide impact.

Solution Exploration

1. Industry Solutions

Research of several companies suggests building a standard capability and product, then customizing on top of the standard. Examples include:

EC : Standardized CRM combined with iPaaS (Tencent HiFlow) for 80 % of needs and a‑PaaS (Tencent Weida) for gap‑filling customizations.

AiYin Interactive : Standardized, modular chatbot built by assembling blocks; extreme customizations are addressed by creating new blocks.

Glodon Menglong : Consolidate projects into a standard product, handle remaining custom needs via business communication or secondary development on the standard base.

What is a Standard? A standard is a unified specification for repeatable items, agreed upon by authorities, serving as a common rule.

Why Standardize? Standards satisfy most client needs and provide strong reusability, reducing system complexity.

Focus of Standards varies by analysis stage: process focus during business analysis, model and capability focus during system design.

1.1 Technical Research

To build reusable capabilities, we surveyed common technical measures and analysis methods. The proposed solution “standard capability + common process + business extension points” meets long‑term business growth and system iteration needs.

Domain Modeling – Stable domain model despite changing requirements; suitable for large, complex projects; advantages: reusable, clear consensus; drawbacks: added complexity and cost.

Business Extension Point Model – Provides extension points at appropriate business nodes; benefits: separates extension logic, improves flexibility; drawback: increased cognitive load with many points.

Micro‑kernel Plugin Architecture – Stable core, plugins add functionality; benefits: extensibility without core changes; drawback: plugin complexity and dependency issues.

Process‑Engine Based Business Orchestration – Uses domain to ensure node stability, process engine for rapid new flow creation; benefits: visual modeling, quick new processes; drawback: heavy engine, learning curve.

All measures aim to separate “invariant” and “variant” parts of the system.

2. Industry Solution Details

2.1 System Background

Insufficient reusable capability and low abstraction of key processes cause heavy implementation for large‑client customizations and high maintenance cost.

Many features and policies lack unified management, raising management cost and slowing client onboarding.

Custom code lacks standards, leading to core logic coupling.

2.2 System Goals

Build standard capabilities and common processes as the foundation, then manage feature strategies centrally and apply lightweight customization on top of the standard.

2.3 Design Highlights

2.3.1 Build Standards

Construction of standard capabilities and common processes improves reusability and reduces maintenance cost.

Standard Capability Construction

Continuous development of pricing, order, and control modules solidifies reusable capabilities, with a dedicated customization module for large clients.

Common Process Construction – Order Example

Abstracted order flow separates core logic from variations, resulting in cleaner links and a unified order model.

Process Abstraction : Pre‑ and post‑customization order chains are isolated; all endpoints adopt the common process.

Link Cleanliness : Logic complexity is confined to the order model, simplifying core calls.

The order model encapsulates over 250 fields, dramatically lowering cognitive cost.

2.3.2 Configuration

Establish a Feature Management Center to unify all features, policies, and decisions, reducing management overhead.

Rule Cache : Locally stores decision elements (features, policies, conditions) to reduce database access.

Decision Result Cache : Caches policy outcomes; unchanged inputs return cached results.

Engine Core : Combines facts and rules to compute policy decisions.

2.3.3 Light Customization

Define allowable customization measures, standardize related logic, and implement limited custom logic on top of standard capabilities and processes.

Extension points are placed in model construction and action execution to realize differentiated implementations.

Deep customizations that do not fit extension points are addressed by creating new processes built on standard capabilities.

COLA extension component supports three‑level extensions (business identity, use case, scenario), e.g., Tmall order – 88VIP scenario – user‑identity verification.

References

ThoughtWorks Modern Enterprise Architecture Framework Whitepaper

TO B essence is “customization” – variation in implementation

Interview with Zhang Xingliang on SaaS as a business model

Avoid reinventing the wheel: audit system case

Travel platform middle‑platform construction

Digital transformation methodology – reading the modern enterprise architecture whitepaper

Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

microkernelprocess modelingstandardizationFeature ManagementEnterprise SaaS
Huolala Tech
Written by

Huolala Tech

Technology reshapes logistics

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.