Making sense of nonprofit email marketing (2027 update)

What to take away

Why 2027 is not 2021

Apple's Mail Privacy Protection, live since September 2021, pre-fetches tracking pixels for Mail app users. Every open those subscribers "record" may be a machine, not a person. Gmail and Outlook followed with their own proxy fetching.

That breaks the metric most nonprofit teams still report to their board. An open rate of 42% might mean 42% of pixels loaded and nothing about who read the appeal.

Click-through rate survives. Reply rate survives. Gift conversion survives. So does the number that actually matters to a program budget: net revenue per delivered address, measured after fees, refunds and the staff time the send consumed.

If your dashboard still leads with opens, the fix is a dashboard change, not a platform migration.

Consent under CASL, and where it expires

Canada's Anti-Spam Legislation governs commercial electronic messages sent from or to Canada. Charitable donation appeals and event invitations generally fall inside it.

Express consent does not expire on its own, but the burden of proof sits with the sender. Implied consent does expire, and the windows differ by recipient type.

Two operational consequences follow. First, a lapsed donor who gave three years ago and never opted in is not a safe send. Second, a re-engagement campaign aimed at that segment needs a fresh consent path, not a bigger subject line.

Record the date, source and wording of every consent. When a donor asks how you got their address, the answer has to be in the CRM, not in someone's memory.

This is not legal advice. Confirm your own consent records and retention schedule with a qualified Canadian charity lawyer or the CRA before you send to a contested segment.

List hygiene and deliverability

Gmail and Yahoo tightened bulk sender requirements in February 2024. Spam complaint rates need to stay under 0.1%, and under 0.3% at the absolute ceiling. Authentication via SPF, DKIM and DMARC became a practical requirement, not a nice-to-have.

A list that has not been mailed in eighteen months is not an asset. It is a complaint generator.

The arithmetic is simple. If your sending domain reputation drops, every future appeal costs more to deliver and lands in worse positions. Cleaning a list of 40,000 down to 26,000 engaged addresses usually raises net revenue even though the send count falls.

Measure net revenue, not gross

Gross revenue per appeal flatters a program. Net revenue tells the truth.

Work it in named variables you substitute with your own figures:

Net revenue per delivered address = (G − F − R − S − C) ÷ delivered addresses.

Run that for every appeal for four quarters. The pattern that usually emerges is that one or two appeals carry the program and the rest consume capacity.

Donor retention is the other half. A program acquiring donors at $85 and retaining 31% after twelve months is buying revenue, not building it. Compare acquisition cost against projected lifetime value before you scale any channel.

Monthly giving and the churn that hides

Monthly giving looks stable on a total-revenue chart and is not. A program adding 200 sustainers a month while losing 160 is growing by 40, and the loss is invisible in the headline number.

Track four figures monthly:

Payment failure is the recoverable slice. Card updater services and a dunning sequence that retries on days 3, 7 and 14 recover a meaningful share of failed payments before the donor ever notices. That work is operational, not creative, and it usually beats a new campaign.

Choosing platforms and agencies without getting sold to

The platform decision is smaller than the agency decision, and both are smaller than the consent and measurement work above.

Pricing changes. Check each vendor's current published rate before you budget, and ask for the renewal terms in writing.

Agency selection has a different failure mode: unverifiable pitches. Ask for the client name, the period, the metric and the method behind any case study. If the agency cannot name the client, treat the number as marketing.

The FTC guidance for marketers using reviews covers fake feedback, selective requests, conditioned incentives and undisclosed paid placement.

Apply those tests to agency testimonials the same way you would to a product review.

For the technology side, the CISA software acquisition fact sheet points buyers toward development practice, supply-chain exposure, deployment and vulnerability management. A donor CRM holds payment data and personal health details in some cases. Ask those questions before signing.

The GOV.UK technology selection guidance recommends adaptable choices, data control, security review and attention to total ownership cost. Platform migration is a three-to-nine-month project, not a switch.

Privacy, PIPEDA and Quebec Law 25

PIPEDA governs how private-sector organizations handle personal information across Canada. British Columbia adds PIPA, Alberta has its own PIPA, and Quebec's Law 25 tightened consent, privacy officer and breach reporting requirements on a phased schedule through 2024.

Bill 96 adds French language obligations for Quebec communications. If you mail Quebec donors, the language question is not optional.

The W3C Privacy Principles statement gives system designers shared privacy concepts and warns against shifting privacy work onto individuals. Apply that to your actual data flow: what you collect, where it sits, who can see it, how long you keep it, and how a donor gets it corrected or deleted.

For US donors giving across the border, IRS treatment and receipting differ from CRA rules. Route those questions to a cross-border giving specialist or the IRS. Do not improvise.

None of this is legal advice. Name your privacy officer, document your retention schedule, and have a qualified lawyer review your consent language before a major appeal.

Sign-off, and why campaigns stall

Most stalled campaigns die in review, not in production. Three causes recur: no named approver, no deadline attached to the approval, and no rule for what happens when the approver is unavailable.

Fix it with a one-page sign-off record per campaign.

A campaign that cannot name its stop authority does not have one.

Worked example: a lapsed-donor reactivation

A mid-size Canadian charity has 18,000 donors who gave at least once but not in 30 months. The team wants to mail all of them.

Step one: check consent. Under CASL, most of that segment has expired implied consent. Mailing them cold is a compliance risk, not a revenue opportunity.

Step two: split the file. Donors who opted in to email at any point stay in the send stream. The rest go to a re-permission campaign with a single clear ask and a one-click opt-out.

Step three: measure. Track reactivation rate, net revenue per delivered address, and spam complaint rate for thirty days. If complaints exceed 0.1%, stop and re-segment.

Step four: decide. If reactivation beats the cost of acquiring a new donor, scale the approach to the next lapsed cohort. If it does not, the file is a suppression list, and that is a legitimate outcome.

The point of the exercise is not to save the list. It is to find out whether the list is worth saving before you spend the send.

Common questions

Do opens still matter at all?

As a diagnostic signal for a specific test, sometimes. As a program metric reported to a board, no. Apple Mail Privacy Protection and similar proxy fetching mean a large share of opens are machine events. Use clicks, replies and gifts.

How often should a nonprofit email a donor?

There is no universal number. The workable rule is to mail at the frequency your engagement data supports and to suppress anyone who has not interacted in twelve months. Frequency that ignores engagement drives complaints and deliverability damage.

What is a safe spam complaint rate?

Google's bulk sender guidance sets 0.1% as the threshold to stay under and 0.3% as the ceiling. Above that, filtering follows. Monitor per-send, not per-quarter.

Can we mail donors who gave years ago but never opted in?

Under CASL, implied consent from a donation expires 24 months after the transaction. Past that window, you need express consent or a fresh consent path. Confirm your specific situation with a qualified Canadian charity lawyer or the CRA.