A donor says “no email.” What should your marketing systems do?

Donor communication preferences should not live in one yes-or-no field. Record the donor's choice by channel and message purpose, preserve where and when it came from, and treat a conflict between systems as a hold on the affected message—not permission to guess.

That model gives a lean team one practical decision: which message can proceed now, and which one needs a named owner to resolve the record first.

Start with channel and purpose

“Do not email” sounds clear until you prepare a receipt, an appeal, and a personal reply. The channel is the same. The purpose is not.

Keep four parts together for every preference:

• Channel: email, text, phone, or postal mail.
• Purpose: the kind of message your organization has defined.
• State: allowed, declined, or unknown under your approved policy.
• Evidence: the donor's wording, source, and effective time.

Do not turn unknown into allowed. Do not interpret a broad request narrowly because a campaign is about to launch. Preserve the request as written, then let your authorized owner apply the organization's policy.

Current nonprofit CRM models already hint at this separation. Microsoft's Fundraising and Engagement contact record includes general contact preferences and a separate receipt-format preference. That product-specific model exposes the underlying issue: “How may we contact this person?” and “How should we deliver this receipt?” are different questions.

Separate donor intent from system state

A donor preference is evidence of intent. An unsubscribe flag, suppressed address, bounced email, or successful sync is system state. Keep both, but do not treat them as interchangeable.

Suppose your CRM says fundraising email is declined, while the email platform still shows the contact on an appeal list. The useful record is not whichever system updated last. It is:

• the donor's current instruction;
• the system values that disagree;
• the messages affected by that disagreement;
• one owner and resolution deadline.

Hold the affected appeal while the owner fixes the sync. A working email address does not erase the donor instruction. A global delivery failure also does not tell you what the donor prefers about postal mail.

This distinction keeps your donor segments tied to explicit entry, exit, action, and suppression rules. A segment can select the right audience and still fail if the sending tool ignores the current suppression evidence.

Define a small purpose map

Use the fewest message purposes that change an operational decision. A lean team might distinguish:

• payment or receipt delivery;
• account or service messages;
• fundraising appeals and newsletters;
• personal relationship outreach.

Those labels are an internal operating map, not legal classifications. The rules for email, text, phone, and mail vary by jurisdiction and context. Have the person responsible for legal or compliance review define what each purpose means, which evidence it requires, and which messages must remain available. The CRM should apply that approved decision, not invent it.

Vendor behavior also differs. WeGive's current donor-portal documentation distinguishes marketing-list settings and a do-not-contact control from transactional payment, receipt, and account messages. Use that only as an example of WeGive's behavior. Your team still needs a written map for its own tools.

The same discipline helps you keep a receipt and a thank-you message in their separate operational roles without assuming they share the same delivery rule.

Work one donor request all the way through

A donor writes: “Please stop fundraising emails. Send my gift acknowledgments by mail.” Do not translate that into a global “no contact” flag, and do not leave it as an inbox note.

Create two instructions:

1. Email plus fundraising: declined, effective at the time of the request.
2. Postal mail plus acknowledgment: preferred, effective at the same time.

Attach the original message or an approved record of its wording. Record the staff member who entered it. Then check the CRM, email platform, receipt workflow, and any active automation that could send one of the affected messages.

Leave unasked questions unknown. The request does not reveal whether the donor wants event texts or a personal email reply. An authorized owner can clarify later if a real message requires the answer.

Use a narrow conflict hold

When connected tools disagree, do not freeze every donor interaction and do not choose the more permissive value. Apply the smallest hold that fully covers the uncertainty:

1. Name the channel and purpose in conflict.
2. Hold messages in that exact scope.
3. Compare the donor evidence and update history.
4. Resolve the source record, sync the connected tools, and verify the final values.

If the conflict cannot be resolved before a send, exclude the donor from that send. The cost is one delayed message. The alternative is letting a stale list override a current instruction.

Make every change explainable

Each preference change needs five audit facts: the instruction, source, effective time, person or process that recorded it, and synchronization result. Keep the prior value in history instead of overwriting the only evidence.

Then test one real contact before a large appeal. Confirm that the source record, segment, sending platform, and final audience all agree. If staff need a new checkout question to capture the choice, apply the same rule used for donation-form custom fields: keep it only when the answer changes a named action with a named owner.