Is a Client Ready for Buyer Intent Data? An Agency Qualification Scorecard

Qualify the operating system, not the appetite for intent data

An agency should approach qualifying clients for intent-data services as a capacity-and-economics decision. “The client wants more leads” is not a qualification standard. A ready client has:

• a defined market with enough relevant companies or people to produce a workable signal pool;
• deal economics that can support data, activation, reporting, and agency labor;
• an agreed ideal-customer profile and exclusions;
• lawful, platform-permitted ways to use each data field;
• an owner for sales or marketing follow-up;
• conversion events that can be measured; and
• realistic expectations about probabilistic signals.

This protects agency growth and gross margin because delivery begins with a bounded use case rather than a vague promise to “find everyone in market.” Intent is evidence of relevant behavior, not proof that a person will buy, has consented to outreach, or should be identified in a message.

Use three outcomes: qualified, qualified for a limited pilot, or not ready. A high total score must never override a hard stop such as an impermissible use, a sensitive-category restriction, no accountable client owner, or economics that cannot cover delivery.

Run a five-stage client-qualification workflow

A repeatable intent data service client qualification checklist needs named inputs, decision rights, and a review cadence.

1. Discovery owner gathers inputs. Record ICP, geography, deal value, gross margin, sales cycle, reachable market, monthly media or outbound capacity, CRM state, current conversion volume, privacy notices, desired channels, and response-time capacity.
2. Data lead tests feasibility. Map candidate topics, estimate signal availability, inspect sample records, define required fields, test identity and enrichment coverage on a representative sample, and document what cannot be resolved.
3. Activation owner checks eligibility. Confirm that each proposed audience or outreach workflow is allowed by applicable law, the client’s notices and choices, contractual data rights, and the destination platform’s rules. Google, for example, says Customer Match uploads must come from a first-party context and imposes account and data-use requirements; a third-party intent record is not automatically eligible for upload (Google Customer Match policy).
4. Commercial owner models the service. Separate wholesale data, setup, media, integrations, QA, strategy, reporting, and contingency. Set a minimum contribution-margin gate and a change-order rule.
5. Approvers decide and record conditions. The agency owner approves economics; the client’s business owner approves goals and staffing; privacy, security, legal, and platform-policy reviewers approve data use; and the delivery lead accepts the SLA.

Review fit before contract, after a bounded test, at the first operating review, and before renewal or material scope expansion. Review sooner if matchability, signal volume, data rights, destination policy, or client follow-up changes.

Approval limits: automation may classify, summarize, draft, and route within approved rules. A human should approve new data sources, new purposes, audience uploads, public or direct outreach, suppression overrides, score-threshold changes, and material spend changes.

Six scorecard components that expose readiness and failure modes

Score each component from 0 to 3: 0 = missing or unacceptable, 1 = material gap, 2 = workable with conditions, 3 = ready. Use the same evidence standard for every prospect. The score is a discussion tool, not a universal benchmark.

1. Reachable-market sufficiency

Test: Define the ICP, geography, exclusions, topic set, expected refresh cadence, and the minimum usable volume required by each activation channel. Ask for a market-coverage sample before estimating delivery.

Pass evidence: a documented addressable set and enough eligible, matched records for the intended workflow.

Failure mode: broad topics create noise, while narrow topics produce too little volume. Platform minimums can make a good source audience unusable; LinkedIn contact lists, for example, must upload at least 300 rows and match at least 300 member accounts before use (LinkedIn contact targeting requirements).

2. Unit economics and budget fit

Test: Model service revenue minus data, integrations, activation labor, reporting, client success, media management, and a rework reserve. Stress-test a lower match rate and slower sales cycle.

Pass evidence: positive contribution margin under the conservative case, with a written owner for overages.

Failure mode: the agency prices from raw-record volume while absorbing strategy, cleaning, support, and campaign work for free.

3. Data readiness and provenance

Test: Inventory first-party records, identifiers, consent or lawful-basis records, suppression lists, CRM duplicates, field definitions, retention rules, and vendor provenance. Test a blinded sample rather than a hand-picked success set.

Pass evidence: a data dictionary, documented source and permitted purpose, stable keys, quality checks, and a deletion/suppression path.

Failure mode: a high match percentage is mistaken for correctness. The UK ICO advises users of data-broker services to perform their own due diligence on collection source, notice, age, consent claims, opt-outs, and rights handling rather than accept an assurance at face value (ICO data-broker guidance).

4. Activation eligibility

Test: Create a field-to-purpose-to-destination matrix. For every route – CRM, ad platform, outbound, nurture, report, or export – name the allowed fields, legal and policy basis, retention, suppression process, and approver.

Pass evidence: at least one valuable activation path is cleared and technically feasible.

Failure mode: the service is sold before discovering that the client cannot upload the audience, use the identifiers, install a tag, or mention the signal in outreach.

5. Follow-up capacity and SLA

Test: Name the queue owner, response window, daily capacity, routing rules, disposition codes, and escalation path. Run a tabletop exercise with sample records.

Pass evidence: the client can act on the expected volume and close the feedback loop.

Failure mode: qualified records age in a spreadsheet while both parties blame signal quality.

6. Evidence and expectation discipline

Test: Define a baseline, eligible population, acceptance criteria, leading indicators, pipeline outcomes, evaluation window, and attribution limits before launch.

Pass evidence: the client accepts that intent is probabilistic and agrees to report accepted, rejected, contacted, meeting, opportunity, and revenue outcomes consistently.

Failure mode: an increase in attributed conversions is presented as incremental impact without a control or suitable experiment.

A prospect with 15–18 points and no hard stop is generally ready for scoped design. Ten to 14 suggests a conditional pilot. Below 10 suggests remediation first. These thresholds are illustrative; calibrate them against your own delivery history and risk tolerance.

Compare discovery, scorecard, and pilot-led qualification

The main alternatives differ in risk, simplicity, and margin protection.

The scorecard should be the default. Use a pilot only to resolve named unknowns, not to postpone basic qualification. State the sample, permitted actions, workload, success criteria, stop date, and what converts into a paid service.

Model pricing from total delivery cost, not signal volume alone

Pricing assumptions should include five cost pools:

1. Data: topic access, records, enrichment, validation, identity services, and overages.
2. Build: market mapping, taxonomy, score rules, integrations, dashboards, and QA.
3. Activation: audience preparation, media operations, outbound preparation, CRM routing, and suppression handling.
4. Governance: privacy, security, legal and platform review, access controls, audit logs, and deletion work.
5. Service: strategy, reporting, client success, optimization, and incident or rework reserve.

Use a simple minimum-economics model:

Required monthly fee = (expected monthly delivery cost + risk reserve) / target delivery-cost ratio

Then compare the fee with the client’s contribution margin per won customer and a conservative range of incremental opportunities. Do not backsolve a guaranteed ROI from an assumed close rate.

BrandWell agency plans range from $2,500 to $5,000 per month, depending on topic count, term, and available contractually scoped topic exclusivity. The current written quote and Order Form control. Treat that as a planning input only; obtain a current written quote and written topic-protection terms before putting it in a proposal. BrandWell’s public quote page confirms that scope varies with market coverage, categories, lead volume, TrafficID, enrichment, routing, and dashboard requirements, but does not display a public price (BrandWell custom quote).

Measure qualification quality before claiming ROI

The best KPIs separate qualification quality, delivery quality, adoption, and business outcomes:

• Pre-sale quality: percentage of prospects with complete evidence; hard-stop rate; time to decision; pilot-to-service conversion.
• Delivery quality: usable-record rate; duplicate rate; required-field completeness; freshness; match acceptance; suppression success; exception rate.
• Adoption: time from signal to review; percentage routed; percentage acted on within SLA; disposition completeness.
• Revenue quality: accepted opportunities, qualified pipeline per eligible record, win rate by signal cohort, realized contribution margin, retention, and expansion.
• Experiment evidence: change versus a predeclared baseline or comparable control, with sample size and uncertainty disclosed.

Calculate agency contribution margin as (service revenue - variable delivery cost) / service revenue. Report it by client and package. Track gross pipeline separately from qualified pipeline, and attributed outcomes separately from incremental outcomes. If an ad platform offers a suitable lift test, use it; Google distinguishes attributed conversions from conversions measured between treatment and control groups (Google Conversion Lift measurement).

There is no honest universal benchmark for acceptable intent-signal accuracy, match rate, or ROI. Establish the client’s baseline, sample method, denominator, and rejection reasons, then improve against that disclosed starting point.

Select clients by fit, not logo value

Best-fit clients usually have a defined B2B or considered-purchase market, sufficient deal value, an identifiable buying committee, approved channels, clean enough systems to accept records, and a team that can respond. Recurring retainers fit clients who want continuous topic monitoring, qualification, activation, reporting, and iteration.

A limited pilot fits a client with attractive economics but one or two testable unknowns, such as topic volume, identity coverage, or destination matchability.

Exclude or remediate clients when:

• the market is too small for the intended channel;
• the product is a low-margin impulse purchase with no credible intent workflow;
• sensitive categories or platform rules make the proposed use unacceptable;
• the client demands claims that reveal private behavior or imply surveillance;
• nobody owns follow-up or feedback;
• the CRM cannot preserve source, suppression, or outcome fields;
• the client requires guaranteed pipeline or revenue; or
• the agency would need to violate its minimum margin or approval policy.

Contract type should follow uncertainty. Use a fixed-scope assessment for diagnosis, a bounded pilot for unresolved feasibility, and a recurring service only after the operating conditions are proven.

Test the full signal-to-evidence chain

The most useful intent-service client qualification test follows the record from source to outcome:

1. Signal source: What behavior occurred, at what unit – person, account, device, or aggregate – and how fresh is it?
2. Identity check: What identifiers support the match? What confidence and ambiguity exist? What remains unknown?
3. Fit check: Does the person or company meet agreed ICP, geography, role, exclusion, and contactability rules?
4. Eligibility check: Is the proposed use permitted by contract, law, notice, user choice, and destination policy?
5. Activation: Which queue, campaign, report, CRM object, or nurture path receives the record? What needs human approval?
6. Evidence: Was the record accepted, acted on, converted, rejected, suppressed, or expired? What baseline or control supports the conclusion?

Do not hide match loss. Google notes that uploaded records may not equal targetable audience size because of user activity, opt-outs, formatting, hashing, and freshness (Google Customer Match troubleshooting). Test volume after every eligibility and match stage, not just at ingestion.

An agent-ready instruction for Claude or ChatGPT can standardize the review:

Given the approved ICP, data dictionary, permitted-use matrix, sample records, and cost assumptions, score each readiness dimension from 0–3. Cite the supplied evidence for every score. Return unknown when evidence is missing. Flag hard stops separately. Do not infer consent, identity, purchase intent, or legal eligibility. Draft recommendations only; do not upload audiences, contact people, change spend, or modify CRM records without the named human approver.

The same approved instructions can optionally be carried out in the browser through Moxby, which is a separate product. That does not make BrandWell a browser product or make Moxby part of the BrandWell plan.

Put margin, data use, and client trust in the risk register

Before signature, record an owner and mitigation for:

• underestimated enrichment, integration, media, or support cost;
• unbounded revisions, custom reporting, or client workspaces;
• wholesale usage spikes and unclear overage billing;
• low signal volume or targetable match loss;
• false matches, stale fields, duplicate records, and identity ambiguity;
• purpose expansion beyond the original data-use review;
• missing lawful basis, notice, consent, suppression, or deletion process;
• ad-platform restrictions or account ineligibility;
• outreach that exposes observed behavior or sounds invasive;
• insecure exports, excess access, uncontrolled retention, or cross-client leakage;
• attribution presented as causation; and
• client inactivity that is mislabeled as provider failure.

For US commercial email, the FTC says CAN-SPAM applies to business-to-business commercial messages and that brands cannot contract away responsibility when another company sends on their behalf (FTC CAN-SPAM guide). Other channels and jurisdictions have different rules. Require jurisdiction-specific legal review rather than treating a checklist as permission.

Qualify for a recurring service, not a one-time list

When an agency sells a recurring buyer-intent service, qualification must cover operating durability:

• enough recurring signal volume, not one impressive sample;
• a refresh, expiry, and suppression policy;
• monthly economics under expected and downside cases;
• named client owners for each action path;
• reporting that includes rejections and unknowns;
• change control for topics, thresholds, destinations, and spend;
• renewal gates based on usable delivery, adoption, pipeline evidence, and client margin; and
• an expansion rule that requires evidence before adding modules.

BrandWell is relevant when an agency wants a separate white-label sales-and-delivery engine rather than a raw list. The planned white-label sales-and-delivery engine supports agency-controlled billing and branded delivery while BrandWell supplies the underlying intent-data workflow on LeadFuze infrastructure. Depending on the use case, the system can support topic reports, fit and identity checks, enrichment, CRM or ad preparation, exports, and agent-ready instructions. Current entitlements, data rights, destination eligibility, and integrations still require written confirmation.

Agencies can purchase BrandWell’s $70 seven-day reseller pilot. It includes agency-branded topic reports and the complete sales playbook under the current written pilot terms. Other product capabilities and any topic exclusivity remain subject to their separate current written scope. Treat it as an acceptance test: predefine the market, topics, sample window, allowed uses, report schema, required human reviews, acceptance criteria, and conversion terms. Conditional topic exclusivity may be available only for certain topics and under written terms; never promise universal exclusivity.

The client is ready when the agency can answer four questions with evidence: Will the market produce usable signals? Can the client legally and operationally act? Can both sides measure what happened? Can the agency deliver at a healthy margin without overstating certainty? If any answer is unknown, scope the unknown into a test – or decline the work until it is resolved.

Before operational use, complete product, pricing, privacy, security, compliance, legal, and platform-policy review.