How to implement B2B attribution

Give your team numbers it can use

As a B2B marketer, you need to know which channels and campaigns deserve the next dollar. That’s difficult when ad engagement, website visits, webinar attendance, and deal data sit in different tools. Attribution connects the recorded journey to pipeline and revenue so you can compare those activities on a shared basis.

A useful implementation gives you a clear path from a report to the accounts, opportunities, and touchpoints behind it. You can explain how credit was assigned, spot missing activity, and make a budget decision with your sales and Finance teams.

Step 1: Agree on the decisions and metrics

Pick a decision you expect to make soon: whether to repeat an event, increase paid search spend, or change the accounts in an ABM campaign. Use that question to choose the data and reports for your first rollout.

Bring the right people together

• Name an owner from marketing, sales, RevOps, and Finance. Marketing defines the decisions; RevOps checks the data and configuration.

• Agree how contacts, accounts, and opportunities relate. Sales and Finance should confirm which stages and value fields represent the outcomes you want to measure.

Define the first rollout

1. Choose two or three decisions for the next planning cycle, such as reallocating paid budget, renewing an event sponsorship, or prioritizing an account segment.

2. Choose 2–3 primary KPIs you will report consistently:
      ◦ Attributed pipeline value by channel: the share of qualified opportunity value assigned under your chosen attribution model.
      ◦ Influenced revenue by campaign: the value of deals with a qualifying campaign touch. Keep this separate from attributed revenue because the same deal can appear in several campaigns.
      ◦ Cost per attributed opportunity: the relevant channel cost divided by the opportunity credit assigned to it, using the same campaign group and measurement window.

3. Define the objects you will measure:
      ◦ Account: how you identify it (CRM Account ID and/or domain)
      ◦ Opportunity: which opportunity types/stages count
      ◦ Touchpoint: what events count (and what does not)

4. Start with the channels and accounts you can measure reliably. List any important gaps and assign an owner to add those sources.

5. Write down the decisions, metrics, included channels, and definitions in one page. Use it as the shared reference when someone asks what a number means.

Include early activity where it is available. For example, a webinar visit that happened before an opportunity was created can help explain how an account first engaged. Starting at the demo request would leave that part of the recorded journey out.

Dreamdata's account-based data model brings go-to-market activity together around companies. It’s the foundation for automated collection and attribution, helping you connect the people involved in a purchase to pipeline and revenue.

Step 2: Audit your stack and tracking for attribution readiness

Audit the sources that matter to your first decision. Check campaign labels, account matches, deal fields, and offline activity. Fix the gaps that would change how you interpret performance.

Assign the data owners

• Choose an owner for UTM naming rules and keep those rules somewhere the campaign team can find them.

• Confirm the CRM fields used for opportunity stage, value, and dates with sales and Finance.

• Agree how events and sales activities will be recorded, including timestamps and the contact or company involved.

Run this audit

1. List the sources you need: CRM, marketing automation or email, website tracking, ad platforms, and relevant offline events.

2. Sample recent campaigns and check their UTM tags:
      ◦ Check for missing or inconsistent utm_source, utm_medium, utm_campaign.
      ◦ Standardize allowed values, casing, and separators.
      ◦ Keep missing values visible in an agreed category, then fix the source of the problem. Avoid guessing a campaign name just to fill a report.

3. Web tracking audit:
      ◦ Check that expected page views and key events, such as demo requests, appear once and under the correct domain.
      ◦ Test how an identified visitor's earlier activity connects to a contact or company. Record what remains anonymous so you understand the coverage.

4. CRM audit (accounts, contacts, opportunities):
      ◦ Account identifier is present and consistent (Account ID and/or domain).
      ◦ Duplicates are quantified (accounts and contacts) with a cleanup plan.
      ◦ Opportunity amount, stage, created date, and close date are reliable.

5. Offline audit:
      ◦ Record attendance and meaningful follow-up with timestamps. Keep event registration and actual attendance distinct.
      ◦ Check that activities connect to the right account. Where an account has several opportunities, agree which deal relationships and stage rules apply.

6. Choose your first integrations from the sources you have checked, then document any missing history or unsupported activity.

Test a company with several contacts. If one person attended an event and another requested a demo, both should appear in the account journey where the available identifiers support the connection. Check subsidiaries and shared domains before combining records.

Dreamdata's B2B tracking records website activity and supports connecting earlier visits to known contacts and companies when identification is available.

Check a familiar journey after setup so you can see which activity is captured and how it is matched.

Step 3: Connect your sources and check the account journeys

Bring the selected sources into a shared account structure. The result should let you move from a campaign to the companies that engaged and the opportunities they reached, with clear rules for unmatched activity.

Choose sources for your first question

• CRM (accounts, contacts, opportunities, revenue, sales activities)

• Marketing automation or email activity relevant to the campaigns you are measuring

• First-party web tracking

• Your main ad platforms and their cost data, when paid activity is in scope

Check the fields and relationships

• A company identifier, such as a CRM account ID or domain, where available

• A person or visitor identifier that supports matching; anonymous activity may not have a known email address

• An event timestamp and a clear time-zone convention

• Channel and source labels with consistent definitions

• Campaign, ad, or content identifiers appropriate to the source

• A separate cost record connected to the relevant campaign, source, and period; costs do not need to be stored on every touchpoint

Validate before sharing the reports

1. Set system-of-record rules:
      ◦ Use agreed CRM records for opportunities and stages, and confirm the value definition with Finance.
      ◦ Document what each integration reads. Configure any audience or CRM write-back separately, with clear ownership of the fields it changes.

2. Define identity resolution rules:
      ◦ How anonymous web visits connect to known contacts/accounts.
      ◦ How you handle multiple domains/subdomains and subsidiaries.

3. Import available historical activity when it is relevant. Confirm what each source retains; a new tracking installation cannot recreate visits that were never recorded.

4. Check a sample of closed-won deals:
      ◦ Start with a manageable sample, including a large deal, a long journey, and a company with several contacts.
      ◦ Compare the deal value, currency, dates, and filters with the CRM. Reconcile attributed credit across the full included journey, including unattributed categories; a marketing-only subset may represent only part of the deal.
      ◦ Confirm touchpoint timelines match known activity.

5. Assign owners for UTM rules, CRM hygiene, integration monitoring, and model changes.

For example, take a $100,000 deal. $60,000 assigned to marketing and $40,000 assigned elsewhere can be consistent with the full deal value. A marketing-only report should show the $60,000.

Investigate unexplained gaps or duplicated credit rather than forcing every filtered report to show $100,000.

Use Dreamdata's Data Hub to define stages, events, and attribution rules, then preview the data those rules select. For example, check that your qualified-opportunity stage includes the same records your sales team expects before using it to compare campaigns.