How accounting firms can use client intelligence to find clients with deep records

What client intelligence means in an accounting firm

Client intelligence is the habit of reading what the firm already knows about each client, such as payroll size, the apps connected to the ledger, how many years of books exist and what changes are coming, to decide where a proactive conversation is worth having. Most firms point it at advisory work: a client adding entities may need a controller, and a client preparing for sale may need quality-of-earnings support.

The same signals can flag a narrower opportunity. Clients with 50+ full-time employees at peak (contractors excluded), years of documented operations and records spread across many business systems are the kind of companies whose operational histories AI labs and data buyers license through SourceX. The firm never needs to open those records to spot the pattern. Counts, dates and system names are enough to decide whether a permissioned conversation makes sense.

The discipline is simple: screen on metadata, ask before you act, and keep tax return information out of it. The accountants partner page explains the program itself; this guide covers the screen.

What you need in place before you screen clients

Four decisions, made once at the firm level, stop every engagement team from improvising.

• An approver. Name the partner, often the managing partner or the risk and quality partner, who signs off on any outside introduction, and record the decision in the client file.
• An attest-client list. Pull the clients for whom the firm performs audits, reviews, certain compilations or examinations of prospective financial information. Referral fee rules are strictest there, and the firm's restricted entity check belongs in the same step.
• An origination rule. Decide how credit works when a staff accountant spots the signal, so the person who noticed is not cut out. The guide to origination credit and outside introductions compares common models.
• A data-use rule. Write down which client fields the screen may use. A short list is safer than a broad one.

How to build a metadata-only client screen in six steps

Use what your CAS, payroll and advisory teams already maintain. Nothing here requires opening a client's files.

1. Export the roster. From the practice management system, pull active US clients with service line, engagement partner, entity count and the year the relationship began.
2. Mark size at peak. Ask the payroll or CAS lead for the highest full-time W-2 headcount in recent years. Leave contractors out and record the peak rather than today's figure, because a company that shrank can still qualify.
3. Count connected systems. Note the apps tied to the general ledger and the tools the client mentions in meetings: CRM, bill pay, expense, payroll, ticketing, project and file-sharing systems. Strong companies often run 10-15+ systems.
4. Record the span. Write down the earliest year of books the firm or the client holds, and whether an older ledger or archived system still exists.
5. Flag the situation. Add any known event: ERP migration, system retirement, sale preparation, ownership transition, acquisition or wind-down. These are the moments when archives get purged and decisions get made.
6. Score and shortlist. Apply the 4S screen below and give each engagement partner no more than a handful of names per quarter.

The 4S screen: size, systems, span, situation

The fourth column is the point of client intelligence for most firms: the same signal that flags records depth usually flags an advisory need too, so one screen feeds both pipelines. A client that reads strong on all four signals is worth raising with the engagement partner. The company fit checker gives a preliminary, non-binding read without contact details, and who qualifies sets out the full baseline, including rights to license the data and an authorized sponsor. Firms with industry niches can prioritize further using the guide to which niche practices hold the deepest records.

Which confidentiality limits apply to client intelligence

Client intelligence is internal analysis; an outside introduction is a disclosure. Treat them differently.

Three sets of obligations usually overlap. Your engagement letters and privacy notice say what the firm will do with client information. Your professional code and state board rules restrict disclosing confidential client information without consent. And if the firm prepares the client's tax returns, Internal Revenue Code section 7216 restricts how preparers use and disclose return information, so keep return data out of any targeting exercise and ask counsel what consent would be needed before using it for anything beyond the return.

Security obligations matter too. The FTC's Safeguards Rule guidance lists tax preparation firms among the non-bank businesses that must maintain an information security program, so a spreadsheet built from client data belongs inside that program, not on a partner's laptop. In a January 2024 staff post (guidance, not a rule), the FTC said that promises not to use customer data for undisclosed purposes are enforceable whether made in privacy policies, terms of service or marketing. That post addresses AI companies, but it is a useful test for any new use of client information: would the client be surprised?

The working rule: the screen stays inside the firm, the first outside step is the client's own permission, and nothing about the client reaches SourceX until the owner agrees. The client can also apply directly at sourcex.si/apply through your referral link.

This is general information, not legal, tax or financial advice. Confirm with your own counsel, tax adviser or professional body before acting.

Common mistakes when firms mine client data for leads

Illustrative example: one quarterly CAS review

Illustrative, fictional scenario. A CAS practice serving about 60 clients runs the 4S screen before its quarterly planning meeting. Three names come out.

• A regional freight broker with 140 full-time employees at its peak, 11 years of books, about a dozen connected systems and a NetSuite migration planned for spring. The engagement partner mentions it to the CEO during the migration kickoff. The CEO asks for details, the firm discloses that it could receive a referral reward, and the company applies through the firm's link.
• An engineering firm that reads strong but is an audit client. The firm leaves it off the list.
• A staffing agency whose most detailed records are candidate personal data. The team parks it until the owner has reviewed privacy obligations with counsel.

Total effort: one agenda item and no client files opened. Firms building a broader advisory pipeline can fold the same screen into their cross-selling conversations.

How introductions and rewards work for a CPA firm

The firm's role ends at the introduction.

1. After the client agrees, a partner or staff member registers as a partner and shares the referral link, or submits the company through the referral form.
2. SourceX qualifies the company on size, history, data breadth and rights.
3. The company builds its own data inventory; the firm does not export, upload or describe records.
4. The company agrees price and terms with SourceX before AI labs and data buyers review the opportunity.
5. The deal closes, the data is delivered under redaction rules agreed in advance, and the company is paid.

Partners earn 25% of the eligible platform fees SourceX actually collects from the referred company's licensing deals, capped at $100,000 per referred company. Rewards become payable only after the buyer pays and SourceX receives its fee; a lead, meeting or signed agreement alone does not trigger payment, and no reward is guaranteed. The reward comes out of SourceX's fee, never out of what the client receives. Credit goes to the first valid referrer whose introduction leads to a verified company application within the attribution window, so agree internally who registers before anyone sends a link.

For CPAs the fee question comes first. Under the commissions and referral fees rule in the AICPA Code of Professional Conduct (ET 1.520), a member in public practice may not accept a commission for recommending a product or service to a client when the firm also performs an audit, review, certain compilations or an examination of prospective financial information for that client, and permitted commissions and referral fees must be disclosed to the client. State boards can be stricter. For what the expansion of client accounting services means for referrals, see CAS growth at Top 100 firms.

Next step

Add the 4S screen to your next quarterly planning meeting and pick one client that reads strong on all four signals. Once the client agrees to explore it, register as a partner and send your referral link.