Pular para o conteúdo

A migração é, antes de tudo, um projeto de qualidade

Também em 19 idiomas

English · Español · 日本語 · Français · Deutsch · Italiano · 한국어 · 简体中文 · Português (Portugal) · 繁體中文 · Bahasa Indonesia · Türkçe · Polski · Tiếng Việt · ไทย · Українська · Español (México) · Français (Canada) · Deutsch (Schweiz)

In a fruit packing shed, crates of sound apples move along the line, and one crate of bruised and split apples is set apart on the floor.
Imagem gerada com IA

Módulo 1 de 5 do capítulo Migração e sincronização, capítulo 10 do curso Data Architect. Cada módulo se sustenta sozinho. Anterior: o capítulo anterior, Integração e movimentação de dados · Próximo: Carregando com volume

A pergunta que este módulo responde: O que deveria acontecer com os dados antes de você começar a carregá-los no Salesforce?

Um projeto de migração pode enganar você. Você tem dados aqui, precisa deles ali, então alguém começa a falar de Data Loader, Bulk API, arquivos CSV, tamanhos de lote e ordem de carga. Essas coisas importam. Não é por elas que eu começo. A minha primeira pergunta é: o que exatamente estamos prestes a mover? Se a origem tem clientes duplicados, endereços inválidos, valores obsoletos, relacionamentos quebrados e quinze anos de dados de que ninguém precisa, o Salesforce vai virar alegremente a nova casa de tudo isso. Carregar mais rápido não resolve. A migração é, antes de tudo, um projeto de qualidade.

Here’s Steve Trucking’s landscape from Checkpoint 1, the weekend RouteDesk retires. RouteDesk is the desktop dispatch tool the coordinators used inside each distribution center (DC) to turn manifests into routes. It holds eight years of delivery history, about 5 million deliveries, and the quality varies by DC. At Checkpoint 1 we decided it retires, and we left one question open: what, if anything, comes into the org? That’s a migration decision, and this is the chapter for it. ⟲

The arrow that runs once Steve Trucking's landscape from Checkpoint 1, in the same places, the weekend RouteDesk retires: people on the left, Salesforce with Field Service in the middle with the Retail Partner portal and the tracking page, the integration layer from chapter 9 to its right, and the systems around it: Snowflake, FleetLink, the finance ERP, photo storage, and RouteDesk, retiring, with 8 years and about 5 million deliveries. One new arrow, in amber, the migration this module zooms in on, carries what's still open from RouteDesk into Salesforce, migrated once. Underneath: one arrow runs once and then disappears; a migration is an integration with an end. {"author": "Steve Simpson", "blueprint": "1.0.1", "builder": "standard 1.54.0", "built": "2026-10-30", "id": "DGM-DA-MG-10", "rights": "\u00a9 2026 Twinshark LLC. All rights reserved.", "template": "system-landscape 0.13.0", "theme": "stevetecharc 1.4.0"} ABOVE: IDENTITY, APPEXCHANGE, WEB PEOPLE RIGHT: INTERNAL SYSTEMS BELOW: CUSTOMER CHANNELS BOTTOM RIGHT: PARTNERS AND PROVIDERS Dispatch coordinators 45, inside 9 partner DCs Delivery crews ~700 on 330 trucks Care and office staff 100: service, ops, finance Retail Partner users ~300 at 3 partners Recipients ~1 million a year Corporate IdP SSO for office staff only Retail Partner portal Tracking page INTEGRATION Snowflake status history, analytics FleetLink raw GPS stays here Finance ERP invoices partners, pays crews Photo storage 4–5 million a year, off core RouteDesk retiring: 8 years, ~5 million deliveries Phone care agents Email email-to-case SMS and email provider day-before message, with a link Redrock REST API, webhooks Keystone CSV over SFTP Brightwave EDI through a third-party network Salesforce Field Service, one US org a link, no login office staff only calls cases in manifests in status out manifests in status out manifests in status out upserts, partner key status events status events latest position only delivery counts delivery counts photos, from the crews' app day-before, with a link what's still open: migrated once Salesforce a community people identity, apps, web an internal system a customer channel the integration layer starts at the system that opens the connection; the label says what moves a shared certificate: trust, no direct call the migration we zoom in on Checkpoint 1's landscape, the weekend RouteDesk retires. One arrow runs once and then disappears: a migration is an integration with an end. © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

One new arrow. It runs once, and then it disappears. ⟲

A migration is an arrow on the landscape like any other. Everything chapter 9 asked of an arrow applies: what moves, how much, by which key, who proves it arrived. The difference is that this arrow has an end. When it’s done, RouteDesk is switched off and the arrow comes off the drawing. That makes it easy to rush, because “we only do it once.” But the data it carries stays in Salesforce for years. ⟲

O sistema legado tem 500.000 registros de clientes, e migramos 500.000 registros de clientes. Tivemos sucesso? Talvez. E se 40.000 forem duplicatas, 75.000 não tiverem atividade há dez anos, milhares tiverem códigos de estado inválidos, alguns não tiverem um identificador utilizável, alguns filhos apontarem para pais que não existem mais e alguns campos significarem coisas diferentes dependendo de quando o registro foi criado? O relatório ainda pode dizer 500.000 na origem, 500.000 no destino, 100 por cento carregado, e o projeto ainda pode ser um fracasso. A completude importa, mas a completude sozinha não prova qualidade.

Então, antes de projetar a carga, quero entender a origem. Que dados existem, e o que significam? Quão antigos, quão confiáveis, quais campos são realmente usados, quais valores são válidos, quais relacionamentos existem, em quais identificadores posso confiar? O que deveria sobreviver à migração, e o que não deveria? É aqui que os capítulos anteriores cobram a conta: o capítulo 4 nos ensinou o panorama da empresa, o capítulo 5 que a qualidade precisa ser projetada, o capítulo 6 a identidade e a autoridade. A migração aplica tudo isso a um conjunto de dados que já existe. Não é só mover linhas. É decidir quais partes do passado merecem fazer parte da arquitetura futura.

Um inventário de migração precisa de pelo menos três respostas: migrar, arquivar ou descartar conforme a política. Um detalhe transacional de cinco anos atrás pode precisar ser guardado por conformidade sem morar no modelo de dados ativo; dados temporários obsoletos podem não ter nem requisito de retenção nem valor; o histórico pode pertencer a um arquivo pesquisável. O princípio de ciclo de vida do capítulo 7 era “projete a saída”, e a migração é o momento perfeito para aplicá-lo. Não carregue todo registro adiante só porque ele está ali.

So zoom in on RouteDesk and take it apart, one class of data at a time. ⟲

Migrate, archive or discard A table of what's in RouteDesk, with a decision for each class of data. Deliveries not yet made, about 6,400 two days ahead: migrate, to Salesforce Work Orders, because the crews need them on Monday morning. Delivered inside the 60-day claim window, about 190,000: migrate, to Salesforce, so a claim's Case links to its Work Order. Delivered and older, about 4.8 million: archive, to Snowflake by the partner key, for claims and five-year reporting. Route plans and coordinators' drafts, eight years of them: discard by policy, once the business confirms nobody must keep them. The decision column is highlighted. Underneath: retiring a system isn't moving every record; know why every class of data moves, or doesn't. {"author": "Steve Simpson", "blueprint": "1.0.0", "builder": "standard 1.46.0", "built": "2026-10-27", "id": "DGM-DA-MG-11", "rights": "\u00a9 2026 Twinshark LLC. All rights reserved.", "template": "system-landscape/inventory 0.12.0", "theme": "stevetecharc 1.4.0"} WHAT'S IN ROUTEDESK HOW MUCH DECISION WHERE IT GOES, AND WHY Deliveries not yet made about 6,400 (two days ahead) Migrate Salesforce Work Orders: the crews need them on Monday morning Delivered, inside the 60-day claim window about 190,000 (60 days) Migrate Salesforce, so a claim's Case links to its Work Order Delivered, older about 4.8 million Archive Snowflake, by the partner key: claims and five-year reporting Route plans and coordinators' drafts eight years of them Discard, by policy Nowhere, once the business confirms nobody must keep them Retiring a system isn't moving every record. Know why every class of data moves, or doesn't. © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Every class of data gets a decision, a place and a reason. ⟲

The deliveries not yet made have to come in: manifests arrive one to two days ahead, so on cutover weekend RouteDesk holds about two days of work, roughly 6,400 deliveries, and the crews need them on Monday morning. The recent history comes in because of claims. A damage claim is a Case in Salesforce, and the agent working it needs the delivery behind it. So deliveries still inside the claim window come in too. Steve Trucking accepts claims for 60 days after a delivery, so that’s about 190,000 more Work Orders. Had the business said 30 days, it would be half that. The window is the business’s number, not ours. ⟲

The other 4.8 million deliveries go to Snowflake, where Checkpoint 1 already sends status history, kept by the partner key so a claim on an old delivery can still find it, and so leadership’s five-year reports still work. They don’t belong in the operational core: that would make Salesforce RouteDesk’s retirement home, and chapter 7 showed what Work Orders at volume cost. And the route plans and the coordinators’ drafts? Eight years of yesterday’s plans have no value once the deliveries are done. But I don’t decide that alone. ⟲

The opposite mistake is “we don’t need the old stuff.” Who decided? Legal, finance, operations, service, compliance? The data architect shouldn’t invent retention policy during a migration; we translate the business and regulatory requirement into architecture. Before excluding data, establish whether it must be kept, for how long, where, who needs access, and whether its relationships and source identifiers must survive. Eight years of Recipients’ names, addresses and phone numbers is exactly that kind of question: archiving them isn’t the same as keeping them forever. The goal isn’t to migrate less. It’s to know why every class of data is moving, or isn’t. ⟲

Test knowledge. Migrate what the new system needs to operate, not everything the old one held. Expect a scenario with years of legacy history: the answer is usually to migrate active and recent records and archive the rest outside Salesforce where it stays reachable, with the retention decision made by the business, not by the migration team. ⟲

Before the steps, the place. All of this preparation happens outside Salesforce, in a working database of its own: Postgres, SQL Server, Oracle, MySQL, even SQLite for a small job, whatever the team knows. Not in Salesforce, and not in a spreadsheet that gets emailed around. I call it the preparation data store, or the preparation store for short. It sits in between: the source feeds it, and it feeds Salesforce. ⟲

The preparation data store sits in between Three cards in a row. On the left, the source: RouteDesk, 8 years and about 5 million deliveries. In between, highlighted: the preparation data store, a database such as Postgres, SQL Server, Oracle or MySQL, where the data is standardized, cleaned, enriched, deduplicated and validated. On the right, the target: Salesforce, one US org. RouteDesk's extract goes into the preparation data store, kept raw; the preparation data store loads the result into Salesforce; and Salesforce sends back what's already there, and the IDs of what was loaded. Underneath: prepare outside, then load the result; the data store holds the source, the steps, and a copy of what Salesforce already has. {"author": "Steve Simpson", "blueprint": "1.0.0", "builder": "standard 1.46.0", "built": "2026-10-27", "id": "DGM-DA-MG-16", "rights": "\u00a9 2026 Twinshark LLC. All rights reserved.", "template": "cells 0.3.0", "theme": "stevetecharc 1.4.0"} THE SOURCE IN BETWEEN THE TARGET RouteDesk 8 years, ~5 million deliveries Preparation data store Postgres, SQL Server, Oracle, MySQL: standardize, clean, enrich, dedupe, validate Salesforce one US org extract, kept raw load the result what's already there, and IDs back Prepare outside, then load the result. The data store holds the source, the steps, and a copy of what Salesforce already has. HOW TO READ IT where the preparation happens © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

The preparation data store sits in between. Every step in this module happens inside it. ⟲

The extract lands there raw. The standardizing, cleaning, enriching, deduping and validating all run there, as code. The quarantine is a table, the counts are queries, and the whole thing can be run again from the top. Some tools connect a database like this straight to Salesforce, copying data down and sending changes back; others export files for a data loader. There are many ways to load data. The principle doesn’t change: prepare it outside, then load the result. ⟲

And it holds more than the source. Pull down the relevant parts of the target org too. Why? Because dedupe has to compare against what’s already in Salesforce, and when the survivor of a merge is a record that’s already there, the new children need its key to link to it. You can’t decide the merge without both sides in front of you. For RouteDesk, the preparation store holds the RouteDesk extract and a copy of what Salesforce already has: the Retail Partner Accounts, the Products and Service Territories, the claim Cases with their delivery references, and the Work Orders the new integration has already created, because the partners don’t stop sending manifests for our cutover weekend. ⟲

Then you practice. Run the preparation, load a sandbox, reconcile, find the rule that was wrong, fix it, run it all again. By the time we reach production, the preparation has run many times, and it’s boring. That’s the goal. ⟲

A minha ordem: padronizar, limpar, enriquecer, remover duplicatas, validar

Seção intitulada “A minha ordem: padronizar, limpar, enriquecer, remover duplicatas, validar ”

When I’m preparing data for migration, I work in this order, and the order matters. All five steps happen in the preparation data store. ⟲

Standardize. Get equivalent values into one form. One source has CA, California and Calif.; another has US, USA and United States. A company comes in as Steve Plumbing, Steve's Plumbing and Steve's Plumbing LLC. Phone numbers arrive as 5555550142, (555) 555-0142 and +1 555 555 0142. Before we compare records or apply matching rules, we need to know which differences are formatting and which are meaningful. ⟲

Names are the classic trap. I had a system with first and last name in one field, and the Salesforce contact needs them apart. “Daniel Reyes” splits easily. Then you get “Mr. Daniel Reyes,” “Daniel Reyes PhD,” “Daniel Reyes III,” and names with two family names, such as Roberto González Suárez, where González isn’t a middle name: split that wrong and you’re in trouble. Build the patterns (two words, three words, a title, a suffix), run the splits, and have the customer validate the results. ⟲

Limpar. Trate o que está errado, inutilizável ou obsoleto: códigos inválidos, datas impossíveis, erros de digitação, valores de preenchimento, campos com dados que não correspondem ao seu significado. Mapeie os valores das listas de opções de propósito: um status de lead legado Cold pode virar Closed no destino. Quando a lista é longa (já vi 200 valores), construa uma tabela de transformação e peça ao negócio que preencha o destino de cada valor de origem. O Salesforce valida as listas de opções com rigor, então pegue esses casos antes que cheguem, e não espere as regras de validação lhe dizerem o que a sua origem contém.

RouteDesk’s “quality varies by DC” shows up right here. Each DC set up its own status codes over eight years, so the same outcome arrives in nine spellings. The transformation table maps every one of them to the Work Order’s status values in Salesforce: ⟲

DCRouteDesk saysWork Order status
DC 1DLVCompleted
DC 4Delivered - OKCompleted
DC 7DONE/INSTCompleted (the install is on its line item)
DC 1NHCannot Complete (nobody home)
DC 4RefusedCannot Complete
DC 7CXLCanceled

The left two columns come from profiling the extract; the right column comes from the business, not from us. Completed, Cannot Complete and Canceled are three of the standard Work Order status values (WorkOrder (abre em uma nova aba), Object Reference, Winter ‘27, checked 2026-10-27). Notice the third row: one RouteDesk code carried two facts, the delivery and the install, and the target model keeps them apart. ⟲

Enrich. Sometimes missing information can be reliably added: a normalized address, a postal code from city and state, a county, a geolocation, a mapping from a legacy code to the new enterprise taxonomy, an enterprise identifier. Enrichment needs provenance. If we add or change information, I want to know where it came from and why I trust it. Some of RouteDesk’s older deliveries don’t name the Retail Partner at all, because each DC served one partner and everybody knew. The DC and the order number’s format tell us which partner it was. That’s enrichment from a rule we can show, not a guess. ⟲

Remover duplicatas. Depois, encontre os registros que representam a mesma coisa do mundo real. Por que não começar aqui? Porque fazer a correspondência de dados sujos é mais difícil: Steve's Plumbing e STEVES PLUMBING LLC ficam muito mais fáceis de comparar depois da padronização. Remova as duplicatas dentro dos dados que estão chegando, depois em relação ao que já está no Salesforce. A regra do capítulo 6 continua valendo: mesclar é destrutivo; vincular não é. Não junte registros porque uma pontuação aproximada parece convincente.

Validate. Finally, test the transformed data against the target: required information present, reference values valid, relationships resolvable, keys unique where they should be, values fitting the model, authority rules preserved. Only then do I want to talk seriously about loading. And don’t expect the import tool to catch everything. Salesforce’s own Trailhead project on the NPSP importer makes the point: it catches invalid picklist values and missing required fields, but if a spreadsheet turned a postal code like 02134 into the number 2134, the importer loads the four digits (Perform an Import Dry Run (abre em uma nova aba), Trailhead, checked 2026-10-27). That’s one more reason the preparation lives in a database with typed columns, not in a spreadsheet. ⟲

Standardize, clean, enrich, dedupe, validate A funnel, top to bottom. RouteDesk's extract, from nine DCs, kept as it came out: the evidence. Standardize: nine DCs' status codes into one list, and one form for phones and addresses. Clean: impossible dates, placeholder phones, products typed by hand. Set aside to quarantine: a delivery with no partner order number, with no key to load it by. Enrich: the Retail Partner, from the DC and the order number. Dedupe: a revised manifest keyed in twice is one delivery, not two. Set aside, merged: the duplicate, with a plan for its stops and notes. Validate against Salesforce: keys unique, products and territories resolve. Set aside to quarantine: a line item whose product doesn't exist in Salesforce. At the bottom, highlighted: load-ready, and a count that reconciles to the extract. Underneath: bad records need somewhere to go, a quarantine with the reason, an owner and a deadline; not silently dropped, not forced in. {"author": "Steve Simpson", "blueprint": "1.0.0", "builder": "standard 1.46.0", "built": "2026-10-27", "id": "DGM-DA-MG-12", "rights": "\u00a9 2026 Twinshark LLC. All rights reserved.", "template": "funnel 0.2.0", "theme": "stevetecharc 1.4.0"} RouteDesk's extract, from nine DCs kept as it came out: the evidence Standardize nine DCs' status codes into one list; one form for phones and addresses Clean impossible dates, placeholder phones, products typed by hand Set aside: to quarantine a delivery with no partner order number: no key to load it by Enrich the Retail Partner, from the DC and the order number Dedupe a revised manifest keyed in twice is one delivery, not two Set aside: merged the duplicate, with a plan for its stops and notes Validate against Salesforce: keys unique, products and territories resolve Set aside: to quarantine a line item whose product doesn't exist in Salesforce Load-ready and a count that reconciles to the extract Bad records need somewhere to go: a quarantine with the reason, an owner and a deadline. Not silently dropped, not forced in. © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Five steps, in order. A record that fails one goes somewhere visible, with its reason. ⟲

Test knowledge. Data quality is fixed before the load, not after. Know the preparation steps (standardize, clean, enrich, deduplicate, validate) and why the order matters: you can’t match records reliably until their values are in one form. Picklist values are mapped with a transformation table the business signs off, because Salesforce rejects values a restricted picklist doesn’t have. ⟲

A tentação é “vamos colocar tudo no Salesforce primeiro e limpar depois”. Não gosto disso como padrão. Assim que os dados ruins estão em produção, participam da produção: os usuários os veem, a automação, os relatórios, as integrações e os agentes os veem, as duplicatas começam a acumular registros relacionados, e a limpeza fica mais difícil. Onde for prático, melhore os dados antes de entrarem no destino. Nem toda imperfeição histórica precisa ser corrigida; é uma decisão de custo-benefício. Mas “a gente limpa depois” não é uma estratégia de migração. É dívida técnica com data de lançamento.

And not every record deserves the same effort. An active customer with open orders, open cases, current contracts and billing relationships is different from one who bought something once nine years ago. Both may need to be kept. For the critical, active data I may require stricter validation, more enrichment, closer duplicate review and detailed reconciliation; for low-value history, preserving traceability may matter more than making every record perfect. Architecture is partly deciding where quality effort creates value. For RouteDesk, the 6,400 deliveries the crews will drive to on Monday get the strictest checks. The 4.8 million going to the archive need their keys and their dates right; nobody will fix eight years of typos in delivery notes. ⟲

Preserve a identidade da origem, e carregue pelas chaves

Seção intitulada “Preserve a identidade da origem, e carregue pelas chaves ”

Um dos erros mais fáceis é perder a capacidade de rastrear um registro do Salesforce até a sua origem. Se o cliente legado CUST-847291 vira a conta 001…, ainda quero saber que ele veio de CUST-847291, para a conciliação, a solução de problemas, as recargas, a auditoria, a correspondência dos registros filhos, a análise de um rollback e as perguntas que chegam depois do lançamento. O novo ID do Salesforce importa dentro do Salesforce; ele não apaga a identidade da origem. É o princípio do External ID do capítulo 9. Migração e integração se encontram na identidade.

At Steve Trucking they meet in one field. Every Work Order already carries the partner’s key, Partner_Delivery_Key__c, an External ID marked unique, holding values like REDROCK-8472914-1 (chapter 9, module 2). RouteDesk kept the partner’s order number on each delivery, so the migration builds the same key and upserts on it, exactly as the integration does. That matters on cutover weekend: if Redrock resends a manifest for Tuesday after we’ve loaded Tuesday from RouteDesk, the integration finds the migrated Work Order and updates it instead of creating a second one. RouteDesk’s own delivery number comes along too, in a plain field, so anyone can trace a Work Order back to the record it came from. ⟲

So don’t rebuild relationships by name (“find the Account whose Name is Acme Distribution”). If the legacy contact carries Legacy_Account_ID = CUST-847291 and the Salesforce Account carries that identifier as an External ID, the relationship is rebuilt deterministically. The Bulk API does this in the load file itself: a column header such as WorkOrder.Partner_Delivery_Key__c says “find my parent by this indexed field,” and Salesforce looks up the parent’s ID for you (Relationship Fields in a Header Row (abre em uma nova aba), Bulk API 2.0 and Bulk API Developer Guide, checked 2026-10-27). Keys turn “I think this child belongs here” into “I know which parent this came from,” and that matters more as the Number of Zeros grows. ⟲

Relationships also create sequence. If contacts reference accounts, the accounts must exist first; if line items reference orders and products, those come first. So migration isn’t one giant load. It’s reference data, parents, children, junctions, then dependent history, in an order the target data model dictates. That’s one more reason the data architect belongs in migration planning: the model tells us how the migration has to be assembled. Here’s RouteDesk’s open work, on Checkpoint 1’s model. ⟲

The model sets the load order Four waves from left to right. Wave 1, already there or reference: Retail Partner Accounts, already in Salesforce; Service Territories, the nine DCs; Work Types: delivery, install, haul-away; and Products, the SKUs and services. Wave 2, the parent, highlighted: Work Orders, upserted by Partner_Delivery_Key__c, each pointing to its Retail Partner by the partner's key, its territory by DC code and its work type. Wave 3, its children: Service Appointments, the scheduled visit and window, pointing to their Work Order by the partner key; and Work Order Line Items, the items on each delivery, pointing to their Work Order by the partner key and to their Product by SKU. Wave 4, what points at it: claim Cases already in Salesforce, updated to point to their Work Order by the partner key. Underneath: parents before children, and every child finds its parent by a key, never by a name; the model sets the order. WAVE 1: ALREADY THERE, OR REFERENCE WAVE 2: THE PARENT WAVE 3: ITS CHILDREN WAVE 4: WHAT POINTS AT IT Products SKUs and services Retail Partner Accounts already in Salesforce Service Territories the nine DCs Work Types delivery, install, haul-away Parents before children, and every child finds its parent by a key, never by a name. The model sets the order. HOW TO READ IT the parent the migration builds around © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Wave 1: what the Work Orders will point to has to exist first. Most of it already does. ⟲

The model sets the load order Four waves from left to right. Wave 1, already there or reference: Retail Partner Accounts, already in Salesforce; Service Territories, the nine DCs; Work Types: delivery, install, haul-away; and Products, the SKUs and services. Wave 2, the parent, highlighted: Work Orders, upserted by Partner_Delivery_Key__c, each pointing to its Retail Partner by the partner's key, its territory by DC code and its work type. Wave 3, its children: Service Appointments, the scheduled visit and window, pointing to their Work Order by the partner key; and Work Order Line Items, the items on each delivery, pointing to their Work Order by the partner key and to their Product by SKU. Wave 4, what points at it: claim Cases already in Salesforce, updated to point to their Work Order by the partner key. Underneath: parents before children, and every child finds its parent by a key, never by a name; the model sets the order. WAVE 1: ALREADY THERE, OR REFERENCE WAVE 2: THE PARENT WAVE 3: ITS CHILDREN WAVE 4: WHAT POINTS AT IT Products SKUs and services Retail Partner Accounts already in Salesforce Service Territories the nine DCs Work Types delivery, install, haul-away Work Orders upsert by Partner_Delivery_Key__c by the partner's key by name of type by DC code Parents before children, and every child finds its parent by a key, never by a name. The model sets the order. HOW TO READ IT the parent the migration builds around © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Wave 2: the Work Orders, upserted by the partner key. ⟲

The model sets the load order Four waves from left to right. Wave 1, already there or reference: Retail Partner Accounts, already in Salesforce; Service Territories, the nine DCs; Work Types: delivery, install, haul-away; and Products, the SKUs and services. Wave 2, the parent, highlighted: Work Orders, upserted by Partner_Delivery_Key__c, each pointing to its Retail Partner by the partner's key, its territory by DC code and its work type. Wave 3, its children: Service Appointments, the scheduled visit and window, pointing to their Work Order by the partner key; and Work Order Line Items, the items on each delivery, pointing to their Work Order by the partner key and to their Product by SKU. Wave 4, what points at it: claim Cases already in Salesforce, updated to point to their Work Order by the partner key. Underneath: parents before children, and every child finds its parent by a key, never by a name; the model sets the order. WAVE 1: ALREADY THERE, OR REFERENCE WAVE 2: THE PARENT WAVE 3: ITS CHILDREN WAVE 4: WHAT POINTS AT IT Products SKUs and services Retail Partner Accounts already in Salesforce Service Territories the nine DCs Work Types delivery, install, haul-away Work Orders upsert by Partner_Delivery_Key__c Work Order Line Items the items on each delivery Service Appointments the scheduled visit and window by the partner's key by name of type by DC code by the partner key by SKU by the partner key Parents before children, and every child finds its parent by a key, never by a name. The model sets the order. HOW TO READ IT the parent the migration builds around © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Wave 3: the children, each finding its Work Order by the same key. ⟲

The model sets the load order Four waves from left to right. Wave 1, already there or reference: Retail Partner Accounts, already in Salesforce; Service Territories, the nine DCs; Work Types: delivery, install, haul-away; and Products, the SKUs and services. Wave 2, the parent, highlighted: Work Orders, upserted by Partner_Delivery_Key__c, each pointing to its Retail Partner by the partner's key, its territory by DC code and its work type. Wave 3, its children: Service Appointments, the scheduled visit and window, pointing to their Work Order by the partner key; and Work Order Line Items, the items on each delivery, pointing to their Work Order by the partner key and to their Product by SKU. Wave 4, what points at it: claim Cases already in Salesforce, updated to point to their Work Order by the partner key. Underneath: parents before children, and every child finds its parent by a key, never by a name; the model sets the order. {"author": "Steve Simpson", "blueprint": "1.0.0", "builder": "standard 1.46.0", "built": "2026-10-27", "id": "DGM-DA-MG-13", "rights": "\u00a9 2026 Twinshark LLC. All rights reserved.", "template": "cells 0.3.0", "theme": "stevetecharc 1.4.0"} WAVE 1: ALREADY THERE, OR REFERENCE WAVE 2: THE PARENT WAVE 3: ITS CHILDREN WAVE 4: WHAT POINTS AT IT Products SKUs and services Retail Partner Accounts already in Salesforce Service Territories the nine DCs Work Types delivery, install, haul-away Work Orders upsert by Partner_Delivery_Key__c Work Order Line Items the items on each delivery Service Appointments the scheduled visit and window Claim Cases already in Salesforce: update the lookup by the partner's key by name of type by DC code by the partner key by SKU by the partner key by the partner key Parents before children, and every child finds its parent by a key, never by a name. The model sets the order. HOW TO READ IT the parent the migration builds around © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Wave 4: the claim Cases already in Salesforce get their link to the delivery. The model sets the order. ⟲

Test knowledge. Keep each legacy record’s key in an External ID field, load parents before children, and let each child find its parent by that key: an upsert on the External ID, and a relationship column in the load file (Parent.External_Id__c) instead of looking up Salesforce IDs. Reloads then update instead of duplicating, and every record traces back to its source. ⟲

Registros mesclados precisam de uma estratégia para os filhos

Seção intitulada “Registros mesclados precisam de uma estratégia para os filhos ”

Eis o problema que as pessoas subestimam. A análise descobre que os clientes legados A, B e C são o mesmo cliente, então o destino recebe uma conta só. Tudo bem. Mas A tem 12 casos, B tem 4 pedidos e C tem 3 contatos. Mescle os pais conceitualmente sem redesenhar o mapeamento dos filhos e você resolveu o pai e quebrou os relacionamentos. Toda decisão de remoção de duplicatas precisa de uma estratégia para os filhos.

RouteDesk has its own version. When a partner revised a manifest on the day, a coordinator sometimes keyed the delivery in again instead of editing it. Now RouteDesk has two deliveries with the same partner order number: one with the old window and two stops, one with the new window and the crew’s notes. Salesforce will have one Work Order, because the partner key is unique. Which window wins, and what happens to the stops and notes on the other one? ⟲

Primeiro, os campos: quando dois registros correspondem, quais valores vencem? Um vazio contra um valor é fácil. Um conflito não: Steven ou Stephen, o e-mail do Salesforce ou o da carga? Escolha um sobrevivente e decida campo a campo, e depois coloque casos de exemplo diante do negócio cedo, para que as digitais dele estejam nas escolhas.

Then the children. I see three options: add the children to the survivor, don’t load the merged record’s children, or load the merged record as inactive, linked to the survivor, with its children kept on it. ⟲

Three ways to handle a merged record's children Three equal panels, three ways to handle the children of a merged record. Add the children to the survivor, one record with everything: at RouteDesk, both deliveries' stops and notes move to the one Work Order; you get everything in one place, simple to use; it costs a merged record's history mixed in; choose it when a wrong merge is unlikely and the history matters. Don't load the merged record's children, a clean record: at RouteDesk, the duplicate's stops and notes stay in the archive; you get a clean record and nothing confusing; it costs that history, gone from Salesforce; choose it when nobody needs that history day to day. Load it inactive, linked to the survivor, an archive of its own history: at RouteDesk, the duplicate loads as canceled, pointing to the survivor, with its stops; you get every child traced and nothing lost; it costs a record that shows up in search and can confuse users, and it needs its own key; choose it when a wrong merge is costly and every child must be traced. Underneath: every merge needs a child strategy; weigh the benefit of joining the data against the risk of joining the wrong records. {"author": "Steve Simpson", "blueprint": "1.0.0", "builder": "standard 1.46.0", "built": "2026-10-27", "id": "DGM-DA-MG-14", "rights": "\u00a9 2026 Twinshark LLC. All rights reserved.", "template": "comparison 0.4.0", "theme": "stevetecharc 1.4.0"} Add the children to the survivor one record with everything AT ROUTEDESK Both deliveries' stops and notes move to the one Work Order YOU GET Everything in one place, simple to use IT COSTS The merged record's history is mixed in CHOOSE IT WHEN A wrong merge is unlikely and the history matters Don't load the merged record's children a clean record AT ROUTEDESK The duplicate's stops and notes stay in the archive YOU GET A clean record, and nothing confusing IT COSTS That history is gone from Salesforce CHOOSE IT WHEN Nobody needs that history day to day Load it inactive, linked to the survivor an archive of its own history AT ROUTEDESK The duplicate loads as canceled, pointing to the survivor, with its stops YOU GET Every child traced, nothing lost IT COSTS It shows up in search and can confuse users; it needs its own key CHOOSE IT WHEN A wrong merge is costly and every child must be traced Every merge needs a child strategy. Weigh the benefit of joining the data against the risk of joining the wrong records. © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Three ways, none free. Notice the third one at RouteDesk: the inactive duplicate can’t reuse the partner key, because the survivor already holds it, so it needs a key of its own. ⟲

Whichever you choose, it happens in the preparation store, before anything loads. That’s where the survivor is picked, sometimes a record that’s already in Salesforce, pulled down with its key. That’s where each child gets the key of the parent it will point to. And that’s where the merged record either keeps its children, hands them over or is left out. Salesforce receives the result of the merge, not the job of doing it. ⟲

Como escolho? Depende de quão críticos são os dados e de se precisamos de rastreabilidade até o que foi carregado. E pese o risco contra o benefício de mesclar, afinal. Juntar o histórico de compras de duas pessoas erradas numa lista de clientes de varejo é um incômodo; mostrar os medicamentos de um paciente no registro de outro é uma violação grave. Quanto mais alto o preço de uma correspondência errada, mais conservadora a mesclagem. Decida o que decidir, cada filho ainda pode ser rastreado até a sua origem? Remover duplicatas numa migração não é “rodar a correspondência de duplicatas nas contas”. O grafo inteiro de relacionamentos importa.

Já estive do outro lado disso. Uma varejista em que eu comprava desde os tempos de faculdade estava migrando os seus dados de clientes e, quando olhamos, Steve Simpson aparecia com uma meia dúzia de e-mails e endereços: o meu endereço da faculdade, a casa que aluguei, os lugares depois disso. Deveríamos mesclá-los? Existem outros Steve Simpsons por aí. O negócio decidiu que juntar o meu histórico antigo não valia o risco de me mesclar com o Steve Simpson errado. Essa é a pergunta a que toda mesclagem precisa responder: qual é o benefício de juntar os dados, e qual é o risco de escolher as pessoas erradas?

Test knowledge. Deduplicate before the load, within the incoming data and against what’s already in Salesforce, with matching rules and a survivor chosen field by field. Expect the follow-up question: what happens to the merged record’s child records? Re-parent them to the survivor, leave them out (and archive them), or keep them on an inactive linked record. Each has a cost. ⟲

Transformações repetíveis, origem bruta preservada

Seção intitulada “Transformações repetíveis, origem bruta preservada ”

O status legado A vira Active, I vira Inactive, S vira Suspended. Onde fica esse mapeamento? Não na memória de alguém, não numa fórmula de Excel copiada de jeitos diferentes em três arquivos, não como um passo manual na sexta à noite. Se recarregarmos amanhã, deveríamos obter o mesmo resultado, então os mapeamentos, as regras de padronização e limpeza, as transformações, as decisões de remoção de duplicatas e a geração de chaves precisam ser reproduzíveis. O processo de migração é código, mesmo quando algumas partes não estão literalmente escritas numa linguagem de programação.

And don’t destroy the starting point. Keep the original extract in raw tables, exactly as it came out, and never change them. Each step writes its result to a new staged table: standardized, cleaned, deduplicated, load-ready. When something looks wrong, I want to know whether it was wrong in the source, changed by our transformation, changed by the load, or changed by Salesforce automation afterward, and with every table kept I can walk back through them and find out. Overwrite your only copy of the source while cleaning it and you lose the evidence. ⟲

I learned how much this pays off on a support migration. A company was moving its support desk from Zendesk into Salesforce, and Zendesk’s tickets were very similar to Salesforce Cases, but not exact. So I loaded every Zendesk export into a SQL database as raw tables, one per Zendesk structure: organizations, users, tickets, custom fields, comments, attachments. Then I built interim tables that got closer and closer to the Salesforce structure, with the field and picklist mappings kept as tables of their own. Then I loaded Salesforce from those tables and brought the Salesforce IDs back. In the end every row had three things: its original Zendesk key, the transformations that had been applied to it, which were never one way, and its Salesforce key. That’s what let me do partial loads and reloads whenever I needed to. ⟲

Here’s the same design for RouteDesk, one step at a time. ⟲

Three sets of keys Four columns from left to right. RouteDesk: deliveries keyed by the RouteDesk delivery number, delivery items by item number, stops by stop number. The preparation store's raw tables, unchanged: rd_delivery, rd_item and rd_stop, each as extracted with its key. The preparation store's tables shaped for Salesforce: sf_work_order, with the partner key built and the status mapped; sf_work_order_line_item, with the product mapped by SKU; sf_service_appointment, with the window and territory; generated from the raw tables, with the mapping tables for status and product, signed off by the business, in amber. Salesforce: Work Orders, highlighted, with the Salesforce ID, the partner key and the RouteDesk number, upserted; Work Order Line Items and Service Appointments, inserted. Last, navy arrows bring each Salesforce ID back to its row in the shaped tables. Underneath: source keys, the transforms, and the Salesforce keys, all kept; now you can reload, load a part, or trace any record back to where it came from. ROUTEDESK PREPARATION STORE: RAW PREPARATION STORE: SALESFORCE SHAPE SALESFORCE Deliveries key: RouteDesk delivery number Delivery items key: item number Stops key: stop number rd_delivery as extracted, with its key rd_item as extracted, with its key rd_stop as extracted, with its key extract extract extract Source keys, the transforms, and the Salesforce keys, all kept. Now you can reload, load a part, or trace any record back to where it came from. HOW TO READ IT a transformation table: the rule, kept as data the Salesforce ID, written back to its row © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Extract: one raw table per RouteDesk structure, each row with its RouteDesk key. These tables never change. ⟲

Three sets of keys Four columns from left to right. RouteDesk: deliveries keyed by the RouteDesk delivery number, delivery items by item number, stops by stop number. The preparation store's raw tables, unchanged: rd_delivery, rd_item and rd_stop, each as extracted with its key. The preparation store's tables shaped for Salesforce: sf_work_order, with the partner key built and the status mapped; sf_work_order_line_item, with the product mapped by SKU; sf_service_appointment, with the window and territory; generated from the raw tables, with the mapping tables for status and product, signed off by the business, in amber. Salesforce: Work Orders, highlighted, with the Salesforce ID, the partner key and the RouteDesk number, upserted; Work Order Line Items and Service Appointments, inserted. Last, navy arrows bring each Salesforce ID back to its row in the shaped tables. Underneath: source keys, the transforms, and the Salesforce keys, all kept; now you can reload, load a part, or trace any record back to where it came from. ROUTEDESK PREPARATION STORE: RAW PREPARATION STORE: SALESFORCE SHAPE SALESFORCE Deliveries key: RouteDesk delivery number Delivery items key: item number Stops key: stop number rd_delivery as extracted, with its key rd_item as extracted, with its key rd_stop as extracted, with its key Mapping tables status and product, signed off by the business sf_work_order partner key built, status mapped sf_work_order_line_item product mapped by SKU sf_service_appointment window and territory extract extract extract generate generate generate status product Source keys, the transforms, and the Salesforce keys, all kept. Now you can reload, load a part, or trace any record back to where it came from. HOW TO READ IT a transformation table: the rule, kept as data the Salesforce ID, written back to its row © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Generate: new tables shaped for Salesforce, built from the raw ones with the mapping tables. The raw tables are still there. ⟲

Three sets of keys Four columns from left to right. RouteDesk: deliveries keyed by the RouteDesk delivery number, delivery items by item number, stops by stop number. The preparation store's raw tables, unchanged: rd_delivery, rd_item and rd_stop, each as extracted with its key. The preparation store's tables shaped for Salesforce: sf_work_order, with the partner key built and the status mapped; sf_work_order_line_item, with the product mapped by SKU; sf_service_appointment, with the window and territory; generated from the raw tables, with the mapping tables for status and product, signed off by the business, in amber. Salesforce: Work Orders, highlighted, with the Salesforce ID, the partner key and the RouteDesk number, upserted; Work Order Line Items and Service Appointments, inserted. Last, navy arrows bring each Salesforce ID back to its row in the shaped tables. Underneath: source keys, the transforms, and the Salesforce keys, all kept; now you can reload, load a part, or trace any record back to where it came from. ROUTEDESK PREPARATION STORE: RAW PREPARATION STORE: SALESFORCE SHAPE SALESFORCE Deliveries key: RouteDesk delivery number Delivery items key: item number Stops key: stop number rd_delivery as extracted, with its key rd_item as extracted, with its key rd_stop as extracted, with its key Mapping tables status and product, signed off by the business sf_work_order partner key built, status mapped sf_work_order_line_item product mapped by SKU sf_service_appointment window and territory Work Orders Salesforce ID, partner key, RouteDesk number Work Order Line Items Salesforce ID Service Appointments Salesforce ID extract extract extract generate generate generate status product upsert insert insert Source keys, the transforms, and the Salesforce keys, all kept. Now you can reload, load a part, or trace any record back to where it came from. HOW TO READ IT a transformation table: the rule, kept as data the Salesforce ID, written back to its row © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Load: the Work Orders upserted by the partner key, then their children. ⟲

Three sets of keys Four columns from left to right. RouteDesk: deliveries keyed by the RouteDesk delivery number, delivery items by item number, stops by stop number. The preparation store's raw tables, unchanged: rd_delivery, rd_item and rd_stop, each as extracted with its key. The preparation store's tables shaped for Salesforce: sf_work_order, with the partner key built and the status mapped; sf_work_order_line_item, with the product mapped by SKU; sf_service_appointment, with the window and territory; generated from the raw tables, with the mapping tables for status and product, signed off by the business, in amber. Salesforce: Work Orders, highlighted, with the Salesforce ID, the partner key and the RouteDesk number, upserted; Work Order Line Items and Service Appointments, inserted. Last, navy arrows bring each Salesforce ID back to its row in the shaped tables. Underneath: source keys, the transforms, and the Salesforce keys, all kept; now you can reload, load a part, or trace any record back to where it came from. {"author": "Steve Simpson", "blueprint": "1.0.0", "builder": "standard 1.46.0", "built": "2026-10-27", "id": "DGM-DA-MG-15", "rights": "\u00a9 2026 Twinshark LLC. All rights reserved.", "template": "cells 0.3.0", "theme": "stevetecharc 1.4.0"} ROUTEDESK PREPARATION STORE: RAW PREPARATION STORE: SALESFORCE SHAPE SALESFORCE Deliveries key: RouteDesk delivery number Delivery items key: item number Stops key: stop number rd_delivery as extracted, with its key rd_item as extracted, with its key rd_stop as extracted, with its key Mapping tables status and product, signed off by the business sf_work_order partner key built, status mapped sf_work_order_line_item product mapped by SKU sf_service_appointment window and territory Work Orders Salesforce ID, partner key, RouteDesk number Work Order Line Items Salesforce ID Service Appointments Salesforce ID extract extract extract generate generate generate status product upsert insert insert ID back ID back ID back Source keys, the transforms, and the Salesforce keys, all kept. Now you can reload, load a part, or trace any record back to where it came from. HOW TO READ IT a transformation table: the rule, kept as data the Salesforce ID, written back to its row © 2026 Twinshark LLC · stevetecharc.com
Clique para ampliar

Keys back: each Salesforce ID is written to its row. Now every row knows where it came from, what was done to it and where it landed. ⟲

So no transformation is allowed to lose fidelity. If a step throws information away (truncates a note, drops a code it didn’t recognize, keeps one of two conflicting values without saying which), I can’t trace back through it, and I can’t repeat the run and get the same answer. Keep the raw tables, keep the staged tables, keep what each load sent and what came back. Then you can rerun the whole thing from the raw extract at any time, and you have the history of every load. This is migration’s version of lineage: preserve where the data came from and what you did to it. ⟲

The preparation store isn’t the only place to stage. In a few special cases, the last stop before the real objects is a plain custom object inside Salesforce: one flat row per source record, no automation, loaded fast. Then a batch process works through the rows and builds the real records. I’ve used it when Salesforce’s own logic had to create the records and was too slow to load through: a scheduling process that took a few seconds per appointment. Module 3 comes back to that case. ⟲

Salesforce uses the same pattern in its own products, and it’s worth learning from. The Nonprofit Success Pack loads a spreadsheet into its NPSP Data Import object first. A dry run checks for matches without creating anything. Then the importer creates the accounts, contacts and donations, and writes each new record’s ID back onto the row it came from (How the Import Process Works (abre em uma nova aba), Salesforce Help, checked 2026-10-27). That’s a reference, not a rule: staging inside Salesforce is the special case, not the way to load everything. ⟲

For RouteDesk, a one-time move with heavy cleaning and merging, the preparation store outside Salesforce does the work, and the load sends the result. ⟲

De 500.000 registros, 497.800 passam na validação e 2.200 não. Descartá-los em silêncio? Não. Forçar lixo nos campos obrigatórios? Não. Parar o projeto até todo registro histórico ficar perfeito? Provavelmente não. Quero um caminho de quarentena e correção que registre qual registro de origem falhou, por quê, qual regra o rejeitou, se precisa ser corrigido antes da virada, quem é o responsável pela decisão e se pode ser adiado. Agora as exceções são trabalho de arquitetura visível em vez de perda de dados escondida.

Before you call a record bad, look at the rule that rejected it. Sometimes the record is fine and the rule is wrong. Salesforce’s Well-Architected framework puts it plainly: “data architecture decisions determine the populations that become visible—and invisible,” and a required email address “excludes populations that don’t have reliable internet access or personal email addresses” (Fairness (abre em uma nova aba), checked 2026-10-27). RouteDesk has thousands of deliveries where the Recipient gave only a phone number. If the new Work Order requires an email, those aren’t bad records. They’re real people the rule would make invisible. Fix the rule, not the person. ⟲

Depois, prove a transformação antes da carga. Com quantos registros de origem começamos, quantos excluímos de propósito, quantas duplicatas foram consolidadas, quantos estão prontos, quantos em quarentena, quantos relacionamentos pai-filho se resolvem, e quais são os totais de controle? Se os números não batem antes da carga, carregar mais rápido não vai ajudar. O módulo 4 prova a migração com contagens, totais, verificações pontuais e relatórios agregados, mas a prova começa aqui.

A IA pode ajudar a limpar. Ela não pode inventar a verdade.

Seção intitulada “A IA pode ajudar a limpar. Ela não pode inventar a verdade. ”

A IA é útil na migração: sugerir mapeamentos, encontrar valores suspeitos, agrupar possíveis duplicatas, identificar anomalias, gerar código de transformação, explicar padrões nos registros rejeitados. Mas se um processo de enriquecimento decide “esta empresa provavelmente fica no Arizona”, provavelmente não basta. E se ele decide que dois clientes se parecem o bastante para serem mesclados, lembre do capítulo 6: uma correspondência falsa pode causar mais estrago do que uma correspondência a menos. Use a IA para acelerar a investigação. Não a deixe fabricar em silêncio dados com autoridade da empresa.

Além do exame: o que a IA muda. A IA pode acelerar drasticamente o perfilamento, o mapeamento, a detecção de anomalias e a correspondência de candidatos durante uma migração. Mantenha regras determinísticas, procedência e revisão humana em torno das decisões que criam, mesclam ou alteram dados de negócio com autoridade.

  • Quais dados da origem são migrados, arquivados ou excluídos, de acordo com o valor de negócio e os requisitos de retenção.
  • Where preparation happens (a preparation store outside Salesforce, a staging object inside it, or both), and which parts of the target org are pulled down to merge against. ⟲
  • Como os dados são padronizados, limpos, enriquecidos, deduplicados e validados antes de entrar no destino.
  • Quais identificadores de origem e qual linhagem sobrevivem, para que cada registro de destino importante possa ser rastreado e conciliado.
  • The load order the target data model dictates, and the keys each child uses to find its parent. ⟲
  • Como as decisões de remoção de duplicatas tratam os campos e os filhos em todo o grafo de relacionamentos, pesadas pela criticidade, pela rastreabilidade e pelo risco de uma mesclagem errada.
  • Quais regras de transformação e de tratamento de exceções precisam ser repetíveis antes de começar qualquer carga em produção.

“Carregamos 500.000 de 500.000 registros” me diz que o carregador funcionou. Quero saber: movemos os dados certos, melhoramos o que precisava melhorar, preservamos a identidade e os relacionamentos, consolidamos as duplicatas com segurança? Conseguimos explicar o que excluímos, reproduzir a transformação, rastrear um registro do Salesforce até a origem e provar o resultado? Faça esse trabalho primeiro. Depois a gente se preocupa com o quão rápido consegue carregar.