Latest industry insights, tutorials, best practices and case studies.
869 articles
2024 年 Temu（拼多多海外）对账逻辑发生「**业务模式两极分化**」的剧变 ——一边是 **Temu 全托管**（商家供货 / 平台代运营代销代发货），一边是 2024 年下半年才大规模放量的 **Temu 半托管**（商家自备货到 Temu 合作海外仓 / 平台负责销售 / 商家自负责售后退货回流 / USD 结算后结汇 CNY）。两种模式对账路径**完全不一样**：全托管走二方核对（无买家订单号），半托管走「**三方两段式**」——平台账单 ↔ 海外仓入库快照 ↔ 商家内部核算 + 资金回流链独立追踪。本文以 2026-08-22 Temu 半托管试点接入定稿记录 + 试点 6 月账期实测为基准，拆透半托管对账新规则 ——核心是 **3 道闸门**（主营收入正循环 SUCCESS / 海外仓入库未映 MISSING_WAREHOUSE / 跨业务类型混淆 ORDER_TYPE_MISMATCH）+ **售后退货回流四件套**（回流单独立记账 + 货品回流登记 + 质检报废标记 + 退运费回拨）+ **USD 结算镜像**（平台收付净额 = 商家 USD 收单合计、商家实收
电商行业涉税面比传统零售宽得多——一家做跨境电商的中型公司，可能同时要面对 6 大税种（增值税、消费税、企业所得税、个人所得税、印花税、关税）的精细化核算，每个税种的计税基础、税率、申报频率、平台代扣场景都不一样。本文把 6 大税种拆成「计税基础 × 平台场景 × 申报节点 × 常见坑」4 个维度逐一拆解，给到一份能直接落到对账系统的税务处理 map——CFO 和税务经理看完，能立刻定位自己公司在哪个税种上还欠着账。
This tutorial centers on DingTalk's topapi/v2/department/listsub open API and consolidates hands-on experience from multiple real integration projects. It systematically covers field semantics, org-tree construction, department mapping with Kingdee Cloud Cosmic, configuration on the Qeasy platform, and common pitfalls, enabling engineers to master the interface in one read.
For B2B retail scenarios, this guide explains how to sync Wangdiantong Other Inbound Orders into Kingdee Xingchen Sales Returns and link them to return request orders. It covers data flow, field mapping, key configuration points in Qeasy, phased implementation steps, and lessons learned.
This guide consolidates the fields, pagination, department filtering, and cross-system mapping practices of the organization user query API, including authentication, request and response structures, extension-field risks, and synchronization governance for CRM user sync, directory maintenance, and permission mapping.
This guide explains how to use Qeasy Data Integration Platform to recursively pull Feishu employee master data by department into the platform data hub, serving as the lookup source for downstream business documents like purchase orders, sales orders, and expense reimbursements. It covers source parameters, field mapping, the write-null operation pattern, and _findCollection/_mongoQuery cross-strategy queries.
Using channel info sync from a reconciliation system as the entry point, this tutorial shows how to configure a single strategy in the Qeasy Data Integration Platform to land JikeCloud channel master data into a MySQL SRM database on an incremental basis. It covers scenario value, field mapping, configuration essentials, scheduling design, and lessons learned, aimed at integration engineers handling supply-chain master data landing.
This tutorial walks through configuring a 'query-only' strategy on the Qeasy Data Integration Platform to periodically pull provincial auxiliary records from Kingdee Cloud. It covers the business scenario, field mapping, platform configuration essentials, phased scheduling, and common pitfalls.
This guide focuses on a single strategy—other inbound orders—covering the full chain from source query, time-window incremental extraction, to target-side field persistence. The focus areas are time windows, status filtering, header/line staged writes, and centralized code mapping management, giving implementation engineers a directly actionable reference.
For supply-chain integration scenarios, this guide explains how to use Qeasy Data Integration Platform to push error and skipped records from specified strategies in the last 24 hours to an O&M group chat via a WeChat Work bot. It covers data flow, field mapping, configuration essentials, and practical lessons learned.
For multi-entity supply chain integration, a nutrition and health brand connected its marketing cloud with a cloud ERP using 66 strategies, automating master data (customers, materials), sales outbound, purchase inbound, return inbound, transfers, and inventory writeback. The solution is organized into four phases, covering code mapping, incremental and full sync, token refresh, and health checks, and can be implemented on the Qeasy Data Integration Platform.
This article focuses on a specific sync strategy—pushing sales orders from JiKeYun into Yongyou NCC as red-letter sales orders (with logistics company) by agent dimension. It walks through scenarios, data flow, field mapping, configuration on Qeasy, phased scheduling, and lessons learned, written for integration engineers working on supply chain scenarios.
This article uses a manufacturing enterprise's production return order synchronization as a case study, breaking down the complete chain: extracting production return data from a MySQL intermediate database, mapping fields through the Qeasy data integration platform, and writing to Kingdee Cloud. It covers data flow, field mapping, platform configuration, scheduling design, and lessons learned.
一篇文章把 v3 重构后的「双轨认证」讲透：网关层（App-Key + scrypt + 路径级 scope + Redis 限流）防止第三方系统越权调用；应用层（JWT + HMAC-SHA256 + x-gateway-user-id 旁路信任链）防止内部 API 被非法访问；业务层（bcrypt 10 rounds + 5 次错密码锁 15 分钟 + IP 60s/10 次限流）防止密码爆破。三个层级相互独立、又通过「同进程可信通道」串成一条完整的防御链。
Stripe 的对账难点不在「charge / refund 怎么识别」，而在「**余额（balance）/ 争议（dispute）/ 储备金（reserve）/ 提现（payout）**」4 套账户各自独立运行、按不同时间窗口结算、对账脚本必须按 `created`（业务发生日）入账而非 `available_on`（资金可用日）入账。本文用 7 张图把 5 类余额活动、5 段争议生命周期、5 条 reserve 释放规则、6 条实操 checklist 讲透——给独立站财务一份「4 账户 × 5 活动 × 5 阶段 × 6 checklist」的完整对账地图。
当月底关账的差异表从 5 天压到 2 天，背后真正的工程支点不是「对账引擎」，而是 23 个差异原因结构化标签码。`diffReason` 给机器读、`diffDisposalSuggestion` 给人读，二者采用**不对称落库语义**——机读码每轮重置确保下游集成脚本精准出单，文本建议保留财务人员手补痕迹不被重跑清空。这套「机读 + 人读」双轨是 v3 对账体系里最容易被忽视、但实际承担 80% 自动化收益的工程支点。
家居大件是和服饰、美妆、3C 并行的"高客单 + 高物流费率 + 多段履约"对账难题行业——服饰物流费率 3-5%、美妆 4-6%、3C 数码 5-8%、食品生鲜 8-12%,**家居大件独有的难题是"送货上楼 + 安装服务 + 退换货大件逆向物流"的复合履约成本分摊——某全屋家居品牌 2026 H1 共发货 87.4 万单、家具订单占比 38.4%、平均客单价 4 287 元、平均物流费率 14.7%、送货上楼占比 71.2%、安装服务占比 28.6%、退货率 11.3%、退货逆向物流费率 32.4%、退货逆向物流成本是正常发货的 2.2 倍** [来源:某全屋家居品牌 2026 H1 物流费运营报告]。**平台账单只显示"扣后净额 = 商品 - 平台券 - 平台扣点 - 物流费(模糊)",内部核算必须把 14.7% 的物流费率拆到"首段干线 + 支线 + 送货上楼 + 安装服务"4 段履约,再按 SPU/SKU 落到主营业务成本**——这就是"按履约段分摊"在家居行业**不是会计技巧、是合规基线**的原因。本文拆解 4 大家居物流场景 × 5 张数据表 × 3 个反直觉发现,讲清楚
This tutorial, grounded in multiple real-world projects, breaks down the "Query Jiandaoyun Material Form Data" API into six parts: scenarios, capabilities, field mapping, Qeasy Cloud configuration, cross-solution best practices, and pitfall retrospectives. Readers can master in one article how this API supports sales-order integration between Jiandaoyun and Kingdee Cloud Cosmic, and how it is encapsulated in the Qeasy Cloud Data Integration Platform.
This tutorial uses a BOM sync failure alerting scenario from a manufacturing enterprise to explain how to use the Qeasy Data Integration Platform to push failed push records from a MySQL interface log table to a DingTalk group robot in real time by business type and responsible person, covering data flow, field mapping, scheduling strategy and common pitfalls.
This tutorial focuses on a specific supply-chain integration strategy: syncing sales order line items from an upstream retail ERP into a downstream analytics warehouse. It walks through data flow, field mapping, configuration on the Qeasy data integration platform, phased implementation, and common pitfalls to help enterprise IT teams avoid rework.
