Why marketing needs the CRM
Marketing can measure a great deal without a CRM: visits, engaged sessions, form submissions, even the campaign that produced each. What it cannot measure is whether any of it mattered. A form fill is a cost; a customer is a result; and the distance between the two, from first conversation to signed contract, is recorded in the CRM or not at all.
That makes the CRM the only place where cost per acquisition (CACCost to acquire one new customer.) can be calculated on real acquisitions, and the only place where lifetime value (LTVTotal value a customer brings over time.) can be attached to the campaign that started the relationship. It is also where marketing and sales either share one definition of a qualified lead or quietly maintain two. A marketing team that treats the CRM as sales' system, and reports on its own numbers from its own tools, ends up optimising towards leads that sales does not want.
The reverse failure is as common. Sales teams build pipelines with stages that mean whatever the last person to edit them thought, and marketing reports on those stages as if they were measurements. The guides in this section are about making the CRM a shared instrument: the fields that have to be filled, the stages that have to be defined, and the integrations that have to flow.
The data model in brief
Every CRM, whatever it is called, is built from a handful of objects joined by relationships. Marketers do not need to administer them, but they need to know what each is and which fields matter.
| Object | What it represents | Why marketing cares |
|---|---|---|
| Contact | A person, keyed on an email address or an identifier | Carries source, consent, lifecycle stage and engagement; the unit most workflows act on |
| Company or account | The organisation a contact belongs to | Needed for business-to-business reporting, account-level scoring and deduplication by domain |
| Deal or opportunity | A potential sale with a stage, an amount and a close date | Where revenue is attributed back to a source; the end of the marketing KPIA metric tied directly to an objective. chain |
| Activity | An email, call, meeting, note or task against a record | Evidence of engagement and of sales follow-up; feeds speed-to-lead reporting |
| Campaign or source | The marketing effort that produced or touched a record | Links contacts and deals to spend; the join that closed-loop reporting depends on |
| List or segment | A saved set of records matching criteria | The audience for a workflow or a send; static or dynamic |
The relationships matter as much as the objects. A deal with no contact cannot be attributed; a contact with no company cannot be rolled up; a campaign that is never associated with a contact never earns credit. The CRM fundamentals guide goes through each object, the properties that should be required, and how source fields are structured so that attribution survives a long sales cycle. Two further guides take the joins that most often break: matching contacts to company records and campaign records in the CRM.
Fundamentals, pipeline, integrations: the core guides
- CRM fundamentals for marketers — the objects and their relationships, properties and required fields, source and attribution fields, duplicates and hygiene, permissions and ownership, and the reports marketing should be able to pull without asking.
- Lead stages and pipeline — written definitions and exit criteria for each stage, the marketing-qualified and sales-qualified hand-off, speed to leadAverage time between a new lead arriving and the first contact from sales. and service levels, stage conversion reporting with a drop-off explorer, and the recycling and disqualification rules that keep the pipeline honest.
- Connecting CRM to ads and analytics — capturing click identifiers and campaign parameters at the form, offline conversion imports, enhanced conversions and the Conversions API from the CRM, closed-loop reporting, and the consent and data-minimisation rules that apply.
Read the core guides in that order the first time. The pipeline guide assumes the fields from the fundamentals guide exist; the integration guide assumes the stages from the pipeline guide are defined, because an outcome sent back to an ad platform is only as meaningful as the stage that triggered it. Scoring and nurture, which act on these records, live in the workflows section, and the lead scoring guide in particular depends on thresholds the pipeline guide sets.
Further guides
Further guides take the jobs that change the record in bulk, the joins between objects, the site changes it depends on, and the routines that keep it trustworthy:
- Moving to a new CRM without losing history — a tutorial for a migration that keeps source, consent and pipeline history: inventory, field and stage mapping, test loads, cutover and switch-off.
- Syncing the CRM and the marketing platform — which records sync, who owns each field, conflict rules, deletions, consent and sync errors.
- Auditing and documenting CRM fields — a tutorial for listing every property, finding what writes and reads it, retiring unused fields, turning free text into picklists and keeping a field dictionary.
- Importing contact lists into the CRM — a tutorial for safe imports: what people were told, field and source mapping, matching existing records, a test sample and workflow enrolment.
- Matching contacts to company records — domain matching, personal addresses, subsidiaries, companies with several domains, ownership and reporting.
- Campaign records in the CRM — one record per effort, names that match the ads, member statuses, adding people automatically, costs and influence reports.
- Reassigning CRM records when someone leaves — a tutorial for the handover: owned records, open deals and tasks, routing groups, emails sent in their name, alerts and access.
- Moving a website without breaking the CRM — a tutorial for a site move or rebuild: the forms, hidden fields, matching and consent settings, workflow triggers and message links the CRM depends on, put in a fixed order and tested with marked records against a baseline.
Data quality and ownership
A CRM decays from the day it is set up. People change jobs, companies rename themselves, forms are added without the fields the others have, and imports bring in records with no source. None of this is dramatic; the effect is that reports drift from the truth a little each month until nobody trusts them.
Quality is therefore a routine with named owners rather than a one-off clean-up. The split that works in most organisations is: sales operations owns the objects, the pipeline stages and the permissions; marketing operations owns the source and consent fields, the forms that populate them, and the lists and workflows that read them; and a named person, not a team, owns the merge queue and the blank-field report. The data quality guide in the analytics hub describes the same discipline for event data, and the principles transfer directly.
Three checks catch most decay. A weekly duplicate review, driven by matching on email and company domain. A monthly report of records missing required fields, split by the form or import that created them, so the gap is fixed at the source rather than patched by hand. And a quarterly review of records with no activity in a long time, which are either recycled into nurture, archived or deleted under the retention rules the organisation has adopted. When workflows depend on these fields, the automation governance guide adds the change-control layer that stops a well-meant field edit from breaking a sequence.
The further guides turn parts of this routine into procedures: the field audit for the fields themselves and the dictionary that records them, list imports that arrive with a real source and a suppression check, reassignment when someone leaves so that no record is owned by a departed colleague, and the sync contract where a second system writes to the same people. Free-text values that drift after the audit can be standardised by workflow, as the field standardisation tutorial shows.
Common CRM problems
Most CRM complaints from marketing resolve to one of a few causes. The table maps each to the guide that deals with it. Symptoms with several causes, such as duplicate contacts or a value that reverts after a sync, have their own diagnosis pages in the troubleshooting section.
| Symptom | Usual cause | Guide |
|---|---|---|
| Cannot say which campaigns produced customers | Source fields blank or overwritten; deals not associated with contacts | CRM fundamentals |
| Marketing and sales report different lead counts | Stages undefined, or defined differently in each team's tool | Lead stages and pipeline |
| Sales says marketing leads are poor | No agreed qualification criteria; hand-off without an acceptance step | Lead stages and pipeline |
| Ad platforms optimise towards form fills, not customers | Won and qualified outcomes never sent back | Connecting CRM to ads and analytics |
| Workflows email the same person twice | Duplicate contacts; no suppression on merge | CRM fundamentals, Automation governance |
| Leads sit untouched for days | No owner assigned at creation; no service level agreed | Lead stages and pipeline |
| Reports change when a field is renamed | No metrics layer; definitions live in individual reports | Metrics layer and definitions |
| Source reports credit leads to "import" | Lists loaded without the real source; a migration that let the import write original source and created dates | Importing contact lists, Moving to a new CRM |
| Opted-out people become reachable again | Suppression check skipped before an import; deletions synced across instead of suppressions | Importing contact lists, Syncing the CRM and marketing platform |
| A corrected value changes back after the next sync | Both systems write the same field; no owner per field; the sync restores corrections | Syncing the CRM and marketing platform |
| Nobody knows which field is the real one | Fields added and never removed; free text where reports need picklists; no field dictionary | Auditing and documenting CRM fields |
| One customer split across several company records | Duplicate companies; personal addresses guessed; one domain held per company | Matching contacts to company records |
| Campaign results cannot be compared with spend | Catch-all campaign records; names that differ from the links; no cost on the record | Campaign records in the CRM |
| New leads go unanswered after someone leaves | The leaver's round-robin slot, sender name and alerts never moved | Reassigning CRM records when someone leaves |
| Leads, sources or workflows go wrong after a website move, rebuild or new form tool | Forms that lose hidden fields or matching, triggers and rules written for the old form or page, and message links to old addresses | Moving a website without breaking the CRM |
| Leads from known accounts or territories land with the wrong seller | A wrong company match, rules read in the wrong order, unmatched territory values or stale rotation groups | Leads assigned to the wrong salesperson |
| Stages, next steps and close dates are weeks out of date | Fields nobody reads, stages without exit criteria, a pipeline reviewed from spreadsheets | Sales not updating the CRM |
The CRM and marketing automation hub places this section in the wider stack, and the conversion tracking expertise page describes how the author approaches the measurement layer these integrations depend on.
