Share the data that connects spend to business outcomes
A marketing agency needs more than ad-platform access. To optimize for revenue, it should receive reliable feedback on which leads were valid, qualified, booked, won, and profitable, plus the service, location, capacity, and economic context behind those outcomes. Access should be role-based, documented, and limited to what the work requires.
Without that feedback, the platform may optimize toward the easiest recorded action rather than the most valuable customer. The agency can lower reported CPL while the business experiences weaker calls, smaller jobs, or lower close rates.
The four data layers an agency should understand
1. Business goals and constraints
Share the outcome the company is trying to change: qualified calls, booked appointments, signed cases, replacement jobs, store sales, or revenue. Add the target markets, preferred service lines, seasonality, capacity, hours, exclusions, and cash-flow constraints.
“Get more leads” is not enough. An HVAC company may need replacements but not maintenance calls. A clinic may have capacity for one procedure and a waiting list for another. A law firm may accept only certain case types or counties. These facts shape keyword, audience, offer, and budget decisions.
2. Acquisition and conversion data
Provide the cleanest available access to advertising accounts, analytics, call tracking, forms, landing pages, tag management, and conversion definitions. The agency should know which actions are primary bidding signals, which are secondary diagnostics, how duplicates are handled, and how consent requirements affect measurement.
Do not share personal passwords. Use platform user roles, business-manager access, and individual accounts with appropriate permissions.
3. CRM and sales outcomes
At minimum, establish shared definitions and recurring counts for:
- New inquiry.
- Valid lead.
- Qualified opportunity.
- Appointment or estimate set.
- Booked job or closed-won customer.
- Revenue and, when feasible, gross profit.
- Disqualification reason, cancellation, or no-show.
Include timestamps and source identifiers where permitted so outcomes can be connected to campaigns. Consistent loss reasons are especially useful: out of area, wrong service, duplicate, price, unavailable appointment, no answer, or unqualified.
4. Economics and value
Share enough unit economics to set rational acquisition targets: average revenue and gross margin by major service line, close rate, refund or cancellation patterns, repeat value, and acceptable payback period. Exact cost detail can remain restricted; the agency still needs decision thresholds.
The minimum viable data set for a local service business
If the current system is fragmented, start with a simple weekly file or dashboard containing lead ID, date, source, campaign, service requested, location, valid or invalid status, qualified status, booked status, revenue when available, and loss reason.
This is not the final state, but it is more actionable than a platform report with clicks and conversions alone. Standardize dropdown values rather than relying on free-form notes. Assign one owner to resolve missing or contradictory records.
What should not be shared by default
Good collaboration does not require unrestricted access to everything. Avoid sharing master passwords, unnecessary customer records, sensitive medical or legal details, full payment data, unrelated employee information, or financial systems beyond the project need.
Use least-privilege access, multi-factor authentication, named users, documented owners, and an offboarding checklist. Confirm how data will be stored, transmitted, retained, and deleted. Regulated businesses should involve their legal, compliance, or security team before sending sensitive data or configuring integrations.
How data improves campaign decisions
Qualified-lead feedback helps distinguish cheap noise from valuable demand.
Booked and closed outcomes reveal which campaigns generate business, not just forms.
Revenue or value signals allow value-based optimization when the platform and data volume support it.
Loss reasons identify whether the fix belongs in targeting, the offer, the landing page, availability, or sales follow-up.
Capacity data prevents campaigns from pushing services or locations the business cannot fulfill.
Google’s Smart Bidding guidance explicitly emphasizes measuring valuable conversion actions and using conversion values when possible. The strategic principle is straightforward: automation learns from the outcomes it can see.
A useful weekly feedback cadence
The business and agency should review a compact funnel from spend to revenue. Compare platform conversions, valid leads, qualified opportunities, bookings, wins, revenue, and gross profit. Break out important service lines and markets, then agree on one or two actions.
Monthly reviews can examine cohort economics, search and audience expansion, creative themes, attribution gaps, and budget reallocation. Quarterly reviews should revisit targets as pricing, capacity, margin, retention, and growth goals change.
Avoid turning reporting into a presentation contest. The goal is a shared source of truth and faster decisions.
Questions to settle during agency onboarding
- Which system is the source of truth for leads, customers, and revenue?
- What exactly counts as a qualified lead and a won customer?
- Who owns tracking, CRM hygiene, call review, and access?
- How quickly will outcome data be available?
- Which service lines and markets have priority or capacity limits?
- What acquisition-cost range is financially acceptable?
- Which personal or regulated data must be excluded or protected?
- How will access be removed when a person or vendor leaves?
Document the answers. Most data disputes are definition and ownership problems before they are dashboard problems.
What a strong agency should do with the data
The agency should translate the information into decisions: suppress weak locations, value high-margin services appropriately, improve query and audience quality, test offers against real objections, and report cost per qualified opportunity or customer alongside platform metrics.
It should also explain uncertainty. Attribution will not be perfect, especially with calls, longer sales cycles, multiple devices, and privacy restrictions. A trustworthy partner distinguishes measured facts from estimates and avoids claiming credit for outcomes the data cannot support.
White Shark Media has managed 20,000+ advertising accounts and optimized 60,000+ campaigns. A focused growth session can help identify which missing business signal is keeping your paid media from optimizing toward qualified calls, booked appointments, and revenue.
Frequently Asked Questions
Does a marketing agency need full CRM access?
Not always. It needs reliable outcome feedback and enough identifiers to connect that feedback to acquisition activity. A limited user role, scheduled export, or controlled integration may be sufficient.
Should I share revenue and profit margins with my agency?
Share enough economics to set acquisition targets and prioritize services. This may be exact values, approved ranges, or relative value tiers, depending on confidentiality and the optimization method.
How often should sales outcome data be sent?
As quickly and consistently as practical. Weekly feedback works for many local-service workflows; longer sales cycles may require staged updates for qualification, opportunity, and closed revenue.
What if the CRM data is messy?
Start with a minimum set of standardized fields and one accountable owner. Do not wait for perfect data, but do not train campaigns on outcomes you know are unreliable.
