Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

CRM & Marketing Automation › CRM for Marketers

Campaign Records in the CRM

How to use campaign records in the CRM: one per effort, names that match the ads, member statuses, adding people automatically, costs and influence reports.
Guideintermediate11 min readUpdated

What a campaign record is for

A contact's source fields say where one person came from. A campaign record says what one effort did. The difference matters as soon as the questions move past "where do leads come from" to "was the open evening worth running", "did the webinar series reach the accounts sales cares about" or "which of last quarter's paid campaigns touched the deals that closed". Those are questions about an effort and everyone it reached, including people who were already in the CRM long before it ran, and only an object that lists its members can answer them.

A campaign record does three jobs. It holds the facts about the effort itself: its name, channel, dates, owner and cost. It lists its members, the contacts it reached, each with a status that says how far they went, from invited to attended or from clicked to submitted. And through those members it connects to the deals their companies went on to open, which is where sourced and influenced pipeline come from. A CRM that does all three can report cost per attending account, pipeline per event and the campaigns present on won deals; one that does none of them leaves marketing reporting on registrations and form fills.

The record does not replace the source fields described in the fundamentals guide. Original and most recent source stay on the contact and are written from the campaign parameters at the form. The campaign record sits beside them, and the two are tied together by using the same campaign name in both places.

One record per effort, named like the ads

The first design decision is what earns a record. The useful test is whether anyone will ever want to filter a pipeline or cost report by it. An event does. A paid campaign with its own budget does. A content offer, a webinar series, a partner promotion or a direct mail drop does; the guide to measuring direct mail lists what a mailing's record should hold, including which customers were held back. An individual email send usually does not; it belongs inside a programme record, with the send-level detail left in the email platform. Too few records and the reports blur everything together; too many and nobody can find the one they need, so the test applies in both directions.

Records usually need a hierarchy of two or three levels: a programme or theme at the top, the efforts within it below, and sometimes a channel split beneath those. Whatever the CRM calls the structure, the aim is that one report can show a whole programme and another can show a single event inside it, without anyone adding up records by hand.

Names carry the join. Google's URL builder documentation, listed in sources, describes how analytics reads campaign parameters from the landing page URL, and the naming conventions guide sets out what each of the UTM parameters should carry: source, medium, campaign, content and term, with a structured campaign name built from the business's convention. The campaign record should use exactly the value that appears in the campaign parameter on the links, so a contact whose source field reads a given campaign can be found as a member of the record with the same name. The fundamentals guide makes the same point about ad platforms: campaign names in the ad account and in the CRM must match, or a mapping table has to translate between them.

The values themselves come from the shared taxonomy. The data operations and taxonomies guide maps each element of the campaign name, such as product line and market, to the CRM campaign field, and the UTM link request SOP is the procedure that checks a requested link against that taxonomy before it is shared. Creating the campaign record is a natural extra line on the same request: the link and the record are made together, with the same name, by the same person.

Record fieldWhat it holdsWhy it matters
NameThe campaign value from the taxonomy, exactly as it appears in the linksJoins the record to source fields and ad platform reports
ParentThe programme the effort belongs toLets one report roll a programme up and another drill into it
ChannelA value from the same controlled list as the medium parameterGroups records by channel without parsing names
DatesStart, end and, for events, the event datePlaces the effort in cohort and period reports
OwnerThe person accountable for the effortSomeone answers for its statuses, cost and closure
CostPlanned and actual spend, entered by a named personTurns members and pipeline into cost per outcome
Status listThe member statuses this record usesKeeps statuses comparable across records of the same kind

Member statuses that mean the same everywhere

A member status records how far each person went. The guide to measuring webinars and events asks for one CRM campaign per event with every person added as a member, using one status list for every event so that reports can compare them; the trade show tutorial and the roundtables and customer events guide apply the same rule to stands and hosted events, with invited, attended and absent recorded.

The principle extends beyond events. Each kind of effort gets one short status list, agreed once and used on every record of that kind, and the statuses read in the same direction: a person moves from a status that means "reached" to one that means "responded" and, for some efforts, to one that means "took the next step". A webinar might use registered, attended, watched the recording and no-show. A paid campaign might use clicked and submitted. A direct mail drop might use sent and responded. What matters is that "responded" means the same kind of thing on every record, so that a report of responders across a programme is not comparing a registration on one record with a meeting on another.

Two habits keep statuses honest. First, a status is never edited backwards: a person who attended stays attended even if they later unsubscribe, because the status records what happened, not what is allowed next. Consent and suppression live on the contact, as the fundamentals guide describes, and every send checks them there. Second, the status list for a record is fixed before the effort starts. A list invented afterwards tends to be shaped by what the data happens to show.

Members added by hand are members that go missing. The reliable pattern is to add people as a side effect of what they do. A form submission whose hidden fields carry a campaign value adds the person to the record with that name, with the status the form stands for. A registration from the event platform adds the registrant, and attendance updates the status after the session. A lead form in an ad platform, connected as the CRM integration guide describes, adds the lead to the record for that campaign. In-person captures, such as badge scans or cards from a stand, are loaded into the event's record by the named person and by the time agreed in the follow-up plan, as the trade show tutorial sets out.

The rule for each entry point is written down: which field carries the campaign value, which status it sets, and what happens when the value matches no record. That last case is the important one. A form that arrives with a campaign value nobody created should land in a review list, not create a new record with a mistyped name and not be dropped silently; the review list is how naming drift is found before it spreads across a quarter's reports. When the source field itself is blank or reads import on a large share of records, the diagnosis for a blank or wrong CRM lead source groups those records by the form, import or system that created them.

Automatic membership is also where imports meet campaigns. A list loaded into the CRM gets the campaign value it was gathered under, and the people on it become members with the status that describes them. The import check in the data operations guide already validates a list's source before it enters the CRM; the campaign value and the member status belong in that same check, decided before the file is loaded rather than after.

Costs, sourced and influenced pipeline

Cost on the record turns a list of members into something the business can compare. Enter the actual spend, including the items that never pass through an ad account, such as venue, travel and staff time for an event, and make one named person responsible for keeping it current. Cost per member, per responder, per attending account and per opportunity can then be read from the CRM, and the acquisition cost on won deals (CAC) can be calculated where the fundamentals guide says it belongs: on deals, not on form fills.

The pipeline reports come from the members' deals. The B2B measurement guide describes the practical design as two rules reported side by side: sourced, where the first touch is recorded on the account, and influenced, where any marketing touch on the opportunity counts. Sourced pipeline gives each opportunity one campaign and can be totalled; influenced pipeline credits every campaign that touched it and cannot, because the same opportunity appears under each of them. The marketing-sourced pipeline entry defines the first, and the disagreement between the two views is information rather than an error. Neither is Attribution in the causal sense, and the guide to choosing an attribution approach explains which decisions each kind of method can support.

Which outcomes the campaign records feed is a measurement plan decision rather than a CRM setting. The measurement plan for lead generation defines conversions at several depths, from form submission to qualified lead, opportunity and won deal, and decides which of them are sent back to the ad platforms. The campaign records should report at the same depths, with the same definitions, so that the CRM and the ad platform describe the same campaign with the same words even when their numbers differ.

Keeping campaign records tidy

Campaign records accumulate like fields and workflows do. Each finished effort leaves a record behind, and within a year or two the list of active campaigns that a form or a seller can pick from is long enough that people pick the wrong one. A light routine keeps it usable. When an effort ends, its owner enters the final cost, checks that the statuses are complete and marks the record as finished, so it drops out of the pick lists while staying available to reports. On a regular cadence, the person who owns the taxonomy looks for records with no members, records with members but no cost, and names that do not follow the convention, and fixes them or closes them.

Two rules protect the history. A finished record is never deleted, because its members and their statuses are part of the pipeline history of every deal they touched. And a record is never renamed to reuse it for a new effort; the new effort gets its own record, so this year's open evening does not inherit last year's attendees. The data operations guide sets the cadence for the wider naming audit, and the campaign record check fits inside it.

Illustrative example — not a real result

Mizzolay Executive Education rebuilds its campaign records

Mizzolay Executive Education, an invented provider of short leadership programmes, has one CRM campaign called "website" with 9,400 members and another called "events" with 2,150. Neither can say which programme a person was interested in. The team creates a parent record for each of its 6 programmes and child records per channel and period, named exactly as the campaign parameter on the links: 31 records for the coming term. Its open evenings get their own records with three statuses, registered, attended and no-show, agreed before the first one runs. Forms now carry the campaign value in a hidden field and add people to the matching record; registrations from the event tool update attendance the next morning. Each record carries its cost, including the £1,800 venue and catering for a typical open evening. At the end of the term the team can report applications per open evening for the first time: the March evening for the finance programme had 64 registered, 41 attended and produced 9 applications, while the online taster had 210 registered, 88 attended and produced 6. Influenced pipeline is read record by record, never totalled.

The example is a small one, but the shape is general: catch-all records are replaced by one record per effort, named like the links, with statuses agreed in advance and members added by the forms themselves.

Common mistakes

  • Catch-all records. A single "website" or "events" campaign holds members but answers no question. Create one record per effort that someone will ask about.
  • Names that differ from the links. A record called one thing and a campaign parameter called another cannot be joined. Use the taxonomy value in both, exactly.
  • A new status list for every record. "Attended" on one event and "showed up" on the next cannot be compared. Fix one list per kind of effort before it runs.
  • Members added by hand after the event. They arrive late, incomplete or not at all. Add people from the form, the event platform and the lead form as they act.
  • No cost on the record. Without it, members and pipeline cannot be compared with spend. Enter actual costs, including venues and travel.
  • Adding up influenced pipeline. Every campaign that touched a deal claims it, so the total exceeds the pipeline. Total sourced pipeline; read influence record by record.
  • Renaming an old record for a new effort. Last year's attendees become this year's. Close the old record and create a new one.
  • Deleting finished records. Their members are part of the history of every deal they touched. Mark them finished instead.

Frequently asked questions

Should every email send get its own campaign record?

Usually not. A campaign record earns its place when someone will ask what that effort produced, which for most teams means a programme, an event, a paid campaign or a content offer, not each individual send. Newsletters and routine sends are better grouped under one record per programme and period, with the individual sends reported inside the email platform. If nobody will ever filter a pipeline report by it, it does not need a record.

What is the difference between a campaign record and the campaign field on a contact?

The field is a label: it says which campaign a person's source parameters named when they arrived, written once for original source and again for most recent source. The record is an object: it can hold its own dates, channel, owner and cost, and it can list every person it touched with a status for each. The field answers where someone came from; the record answers what an effort did, across everyone it reached. They should use the same name, so one can be found from the other.

Can influenced pipeline be added up across campaigns?

No. Influenced pipeline counts an opportunity for every campaign that touched any of its contacts, so the same deal appears under several campaigns and the totals add up to more than the pipeline that exists. Read influence campaign by campaign, or as a share of opportunities touched, and use sourced pipeline, where each opportunity has one first touch, when a total is needed. The B2B measurement guide recommends reporting the two side by side for exactly this reason.

Terms used on this page

Frameworks

Where this page is used

Also related

Sources

  1. [GA4] URL builders: Collect campaign data with custom URLs — Google Analytics Help

Related