CRM Attribution Agency: 10 Questions to Ask Before You Hire

Written by: Aaron Rovner, Founder, Saas Hero

Key Takeaways

• CRM attribution sends offline conversion data from Salesforce or HubSpot into Google Ads and LinkedIn so algorithms chase qualified pipeline instead of simple form fills.
• Most agencies that claim CRM attribution still run MQL attribution, while real practitioners separate primary and secondary conversions and push lifecycle-stage events as bidding signals.
• Technical requirements include GCLID capture, lifecycle-stage mapping, UTM and CRM hygiene ownership, server-side tracking, and Consent Mode v2 compliance across the full data architecture.
• Account-level attribution rolls all contacts in a buying committee into one account record, unlike contact-level models that credit only the form submitter.
• SaaSHero is a demand generation agency that delivers CRM attribution meeting every technical requirement outlined in this article.

See How SaaSHero Handles CRM Attribution

Why CRM Attribution Matters Now

The paid acquisition role changed when platforms took over most levers. Smart Bidding sets the price, broad match decides which queries qualify, and Performance Max chooses the inventory. Human control now focuses on which conversion events the algorithm pursues and how closely those events track to revenue.

An algorithm pointed at a form fill finds the people most likely to fill out forms, such as students, job seekers, competitors, and existing customers, while reporting a falling cost per conversion. This is a self-fulfilling prophecy, and the mechanism is the platform succeeding at the goal it was given. The problem sits in the goal, not in the machine.

The measurement layer broke separately from that shift. Third-party cookie restrictions, browser tracking prevention, consent requirements, and cross-device journeys each removed part of the path between a first impression and a signed contract. The click appears in Google Ads or LinkedIn. The opportunity appears in Salesforce or HubSpot months later. Nothing connects them unless someone builds and maintains the join.

Boards and private equity operating partners now ask in finance language about CAC payback, pipeline coverage, and which spend produced qualified pipeline this quarter. Most reporting stacks cannot answer those questions. Most agencies claiming CRM attribution are doing the same MQL attribution described earlier, just with better branding.

Get A CRM Attribution Readout On Your Current Spend

What CRM Attribution Actually Requires From Your Agency

CRM attribution runs on a specific data architecture, not on a reporting style. Each component needs a clear owner and a working connection to the rest of the stack.

Offline Conversion Import. CRM events must flow back into Google Ads and LinkedIn as conversion signals. For Google Ads, this means capturing the GCLID at click time and storing it in the CRM as a custom field. The conversion is then uploaded with fields including conversion action name, conversion time, and conversion value. Google Ads Data Manager allows direct connection of HubSpot and Salesforce as data sources, mapping lifecycle or opportunity stages to Google Ads conversion actions. For LinkedIn, the equivalent mechanism is the LinkedIn Conversions API. An agency that cannot describe these mechanisms by name is not doing offline conversion import.

Lifecycle-Stage Mapping. A sales-qualified lead or closed-won event becomes a bidding signal only when lifecycle-stage changes in the CRM are configured to flow back to the ad platforms. Google Ads Data Manager’s HubSpot connector maps HubSpot lifecycle stages such as MQL and SQL to Google Ads conversion actions through OAuth authorization, field mapping, and configuration of which deal stages trigger uploads. Salesforce uses the Campaign Influence object and opportunity stage mapping via Google Ads Data Manager to connect CRM milestones to conversion actions. The plumbing differs by stack, and many agencies default to HubSpot-only experience.

Primary Vs. Secondary Conversion Distinction. Google Ads conversion actions for offline CRM milestones should start as Secondary and remain Secondary until they have delivered complete, stable data for at least one ordinary conversion cycle. Only then should the stage with enough volume and the strongest revenue relationship become the Primary bidding action. Secondary conversions remain visible in reporting but stay excluded from account-wide optimization. Content downloads and webinar registrations indicate interest, but they do not indicate a buyer. Treating them as bidding signals trains the account toward the wrong audience.

UTM And CRM Hygiene Ownership. Incomplete UTM tagging is the most frequent attribution failure point, because paid traffic without proper UTM parameters usually falls into direct traffic or unattributed sessions. Someone must own the naming conventions, validate them continuously, and ensure GCLIDs are not stripped by redirects or lost when a lead converts to a contact in Salesforce.

Server-Side Tracking And Consent Mode. Server-side tracking has replaced pixel-based tracking as the default in B2B attribution stacks, driven by iOS privacy changes, browser cookie deprecation, and ad-blocker prevalence. Google’s Consent Mode v2 enforcement took effect in March 2024 and made consent-aware measurement mandatory for any B2B brand running paid media into European markets. An agency still relying on 2021-era pixel attribution runs on a broken measurement layer.

Who Configures Each Piece. Google Tag Manager handles the tag layer, the CRM team owns field mapping and lifecycle definitions, and the marketing automation platform such as HubSpot, Marketo, or Pardot governs the handoff to sales before the ad platforms receive the signals. This chain is exactly the work most agencies scope out, which leaves the client coordinating between parties who each own one piece while nobody owns the full path.

For a deeper look at how revenue-focused agencies use CRM data operationally, see How Revenue-Focused Agencies Use Your CRM To Drive Growth.

MQL Attribution Vs. Closed-Won Attribution: The Difference That Matters

The technical requirements above explain how CRM attribution works. The next step is understanding how it differs from the MQL attribution most agencies actually deliver.

The key difference is the data source and trigger event: standard conversion tracking fires on a page event such as a form submission, while SaaSHero’s CRM-connected approach pushes lifecycle stage events, including when a lead becomes a sales-qualified lead, an opportunity is created, or a deal closes, back into the ad platforms as the optimization signal.

The same self-fulfilling-prophecy mechanism described earlier also runs in reverse. An algorithm optimized toward form fills finds the cheapest people to convert, such as students, job seekers, competitors, and companies below the ICP floor. Cost per lead falls, lead volume rises, the dashboard improves in the metrics the board sees, and the pipeline the sales team can work stays flat. Each month this continues, the bidding model becomes more accurate at finding the wrong people.

In a multi-month B2B cycle with a buying committee, a mis-specified conversion event trains the account toward the wrong audience for a full quarter, and the CRM shows the damage only after the budget is spent. Mid-market B2B SaaS deals close in roughly 84 days on average, while enterprise deals take 170 days or more. These timelines make the choice of conversion event a strategic decision made before launch rather than a reporting preference applied afterward.

MQL-based reporting fails the CFO test on predictability, comparability, and actionability. MQL-to-revenue conversion varies too widely to model, no standard unit cost enables cross-functional comparison, and volume metrics do not indicate optimal resource allocation.

How To Vet An Agency’s Attribution Claims In One Discovery Call: 10 Questions

The questions below separate real practitioners from rebranded lead-gen shops. Run them on your current agency before your next board meeting.

1. Are You Optimizing Campaigns Around CRM Data Or Just Form Submissions? A real practitioner describes offline conversion import and lifecycle-stage feedback. A pretender talks about lead quality scores or form optimization.
2. Which Conversion Events Are Used For Account-Wide Optimization, And Which Are Tracked But Excluded? A real practitioner names primary vs. secondary conversion architecture. A pretender cannot distinguish them or says all conversions are equal.
3. How Do Lifecycle-Stage Events Get Returned To The Ad Platforms? A real practitioner describes Google Ads offline conversion import, LinkedIn Conversions API, or native CRM connectors. A pretender says they use UTM parameters or GA4 goals.
4. Who Owns The Landing Page The Campaign Points To? A real practitioner owns design, build, and testing. A pretender defers to the client or a web team.
5. What Does The Monthly Report Lead With: Leads And CPL, Or Pipeline And CAC Payback? A real practitioner leads with pipeline created by channel, cost per SQL, and CAC payback. A pretender leads with impressions, clicks, CPL, and impression share.
6. What Happens To The Tracking Configuration If We Leave? A real practitioner confirms the client owns all accounts, assets, and configurations. A pretender is vague or says configurations are proprietary.
7. How Do You Handle Buying Committees Where Six People From One Company Touch The Funnel? A real practitioner describes account-level attribution. A pretender uses last-click, which credits whichever contact converted.
8. What Is Your Process For UTM And CRM Hygiene? A real practitioner has documented ownership and validation. A pretender assumes the client handles it.
9. How Do You Handle Consent Mode And Server-Side Tracking? A real practitioner describes a current implementation. A pretender is still relying on third-party cookies.
10. Can You Show Me A CRM-Connected Dashboard From A Current Client? A real practitioner shows a live Looker Studio or HubSpot dashboard with pipeline by channel. A pretender produces a PDF of platform metrics.

For a complete framework on what agency reporting should look like, see Demand Generation Agency Reporting: The Complete Guide.

Use These 10 Questions On Your Next Discovery Call

Stack-Specific Breakdown: HubSpot, Marketo, And Pardot

The vetting questions above apply to any stack, but the answers depend on which marketing automation platform you run. Most agencies assume HubSpot, yet the plumbing differs materially across stacks and those differences affect what is possible in lifecycle-stage feedback and offline conversion import.

HubSpot + Salesforce. Google Ads Data Manager’s HubSpot connector maps lifecycle stages to conversion actions via OAuth authorization, field mapping, and configuration of which deal stages trigger uploads. HubSpot’s attribution reports support first-touch, last-touch, linear, U-shaped, W-shaped, and full-path models natively at Marketing Hub Enterprise. When Salesforce is the CRM of record alongside HubSpot as the marketing automation platform, lifecycle definitions must stay synchronized between both systems or attribution breaks at the handoff.

Marketo + Salesforce. Marketo Engage has a native LaunchPoint integration that automatically uploads offline conversion data from Marketo to Google AdWords, using a one-directional push where conversion data appears only in the Google AdWords portal, not in the Marketo UI. Lifecycle-stage events can also pass through Salesforce Campaign Influence or through a middleware layer. Google Ads Data Manager connects one Salesforce object per connection, so importing events from both Lead and Opportunity objects requires separate connections. A common failure is losing the original GCLID when a Lead converts, because the custom click-ID field is not mapped forward to the resulting Contact or Opportunity.

Pardot + Salesforce. Pardot’s campaign influence model relies on Salesforce Campaign Influence, which is not enabled by default and requires custom reporting for multi-touch attribution. Native Salesforce Reports do not calculate multi-touch attribution automatically, so external logic is required for U-shaped, W-shaped, and time-decay models. An agency that has only worked in HubSpot will not know where Pardot’s attribution gaps sit.

For a detailed treatment of multi-channel attribution across enterprise stacks, see Enterprise Demand Gen Multi-Channel Attribution Guide.

Account-Level Vs. Contact-Level Attribution For B2B Buying Committees

Gartner’s B2B buying journey research found that a typical buying group for a complex B2B solution includes 6 to 10 decision-makers, each gathering 4 to 5 pieces of information independently before sharing findings with the group. Contact-level attribution assigns touchpoints and outcomes to one identified person. In a committee-driven deal, that means crediting whichever contact submitted the form, often a researcher or junior evaluator, while the CFO, COO, and procurement lead who influenced the decision remain invisible.

Account-level attribution aggregates all touchpoints from all contacts at a company into one account record and attributes pipeline or revenue at the account level. Implementing account-level attribution requires all contacts at target accounts to be tagged to the account record in the CRM, all marketing touchpoints for those contacts to be logged, account engagement scoring that aggregates individual contact engagement into an account-level buying signal, and deal attribution connected to the account record rather than to any single contact.

Cometly is a category example of an account-level attribution tool built specifically for B2B SaaS, connecting ad platforms, CRM data, and pipeline in a unified view. The distinction between contact-level and account-level attribution defines whether you measure one person’s journey or the full deal.

What A CRM-Attributed Report Actually Shows

The contrast between a platform-metrics PDF and a CRM-connected dashboard is the clearest diagnostic for whether an agency runs MQL attribution or closed-won attribution. The table below highlights the four report elements that differ most clearly between the two.

Multi-touch attribution is the model long B2B cycles support. Last-click falls away because the data no longer fits it. Without multi-channel attribution, teams tend to over-invest in last-click channels like branded search or direct traffic, while upper-funnel activities like brand campaigns appear ineffective because they rarely get the final click.

Why SaaSHero Is The Standard To Hold Every Agency To

SaaSHero is the demand generation agency with CRM attribution that the vetting rubric above is designed to surface. Every technical requirement described in this article sits inside SaaSHero’s standard operating method rather than as an add-on or an enterprise tier.

The method in practice includes the following elements:

• Primary vs. secondary conversion hierarchy configured in every account, with only primary conversions driving account-wide bidding
• Lifecycle-stage events pushed back into Google Ads and LinkedIn so the algorithm learns from qualified pipeline outcomes instead of form fills
• CRM-connected Looker Studio and HubSpot dashboards showing pipeline created by channel, cost per SQL, and CAC payback, which matches the vocabulary boards use
• Landing pages owned end to end through design, build, hosting, and A/B testing in Unbounce, so the post-click experience stays in scope and the highest-leverage conversion variable does not sit in a web team’s backlog
• Flat retainer indexed to total monthly ad spend rather than channel count, so reallocation carries no fee consequence and channel-mix recommendations rest on evidence alone

Verifiable credentials include the following:

• Founded 2018, with eight years operating in B2B SaaS paid acquisition
• Google Premier Partner, which represents the top 3% of Google Partners
• G2 High Performer in Digital Marketing for more than two consecutive years, ranked #20 of approximately 6,000 agencies
• More than $60M in lifetime ad spend managed across 100+ B2B companies
• Approximately 20 full-time in-house specialists, with no outsourced execution

For context on how SaaSHero’s pricing model compares to percentage-of-spend and per-channel arrangements, see Best Demand Generation Agency Pricing Models For B2B SaaS. For the distinction between lead generation and demand generation that underlies the attribution question, see Lead Generation Agency Vs Demand Generation: B2B SaaS.

Talk With SaaSHero About Your Attribution Setup

Frequently Asked Questions

The questions below address the practical concerns that usually come up once the rubric and stack breakdown above have been reviewed.

How Long Before CRM Attribution Produces Readable Data?

For post-migration HubSpot activity, pipeline attribution data should be trustworthy within 30 to 60 days if the migration was executed correctly. The 90-day mark typically provides a full quarter of clean data. More generally, pipeline attribution timing depends on sales cycle length, as noted earlier when discussing long B2B cycles. Leads and qualified leads appear within days, while pipeline attribution appears within the first weeks as opportunities open.

The first 30 days focus on setup and validation. During that period, conversion tracking is rebuilt, CRM integrations are configured, and the primary vs. secondary conversion architecture is established. Optimization signals begin accumulating in month two. A board-ready view of pipeline by channel generally requires enough data to cover the sales cycle, which is why engagements use a validation gate at 90 days rather than a judgment at day 45.

Do We Need To Change Our CRM?

You can keep your current CRM. The work sits in configuring the connection between your existing CRM and the ad platforms, not in replacing the CRM. Whether you run HubSpot, Salesforce, Marketo, or Pardot, the mechanism stays consistent, because lifecycle-stage events and closed-won data are mapped to Google Ads conversion actions and the LinkedIn Conversions API.

The CRM remains the system of record. The change occurs in what the ad platforms are instructed to optimize toward.

What If Our Agency Says They Already Do This?

Run the discovery call questions above. If the agency cannot describe offline conversion import by name, cannot distinguish primary from secondary conversion architecture, and cannot explain how lifecycle-stage events return to the ad platforms, they are still running MQL attribution with updated branding. Ask to see a live CRM-connected dashboard from a current client.

A platform-metrics PDF does not qualify as the same thing. The questions are designed so that only teams with real technical implementation can answer them clearly.

What Is The Difference Between Marketing-Sourced And Marketing-Influenced Revenue?

Marketing-sourced revenue refers to opportunities where marketing generated the first meaningful engagement that brought the account into the pipeline. Marketing-influenced revenue captures any opportunity where marketing touched the account during the buying journey, regardless of who initiated the relationship.

Both should be tracked in a closed-loop measurement system, and they must be reported as separate line items. Conflating them inflates marketing’s apparent contribution and gives finance teams grounds to question all reported figures. Board-ready reporting leads with sourced revenue as the conservative floor and layers influenced revenue with a transparent methodology.

How Does Account-Level Attribution Differ From What Most Agencies Deliver?

Most agencies operate at the contact level and credit the channel that touched the individual who submitted the form. In a B2B deal involving six to ten decision-makers, that approach hides the CFO who read comparison pages, the VP of Engineering who attended a webinar, and the procurement lead who reviewed a case study.

Account-level attribution rolls every tracked touchpoint from every known contact at a company into one account record and attributes pipeline and revenue at the account level. It requires all contacts to be tagged to the account record in the CRM, all marketing touchpoints to be logged, and deal attribution connected to the account rather than to any single contact. Many agencies skip this work because it requires owning the CRM integration instead of only the ad account.

Conclusion: Run The Questions Before Your Next Board Meeting

The vetting rubric in this article is designed for a single discovery call. Ask your current agency which conversion events drive account-wide optimization, how lifecycle-stage events return to the ad platforms, who owns the landing page, and what the monthly report leads with. The answers reveal whether you are working with a CRM attribution practitioner or an MQL shop with updated branding.

SaaSHero is the standard every agency in this category should be held to and the worked example of what the answers above look like when the implementation is real.

Schedule A Call To Audit Your Attribution