Display costs $86 to acquire an email address and returns $1.11 — the published ad table is a ranking, not a menu

Nonprofit digital advertising is usually planned as a mix, with each channel given a share of a budget that was set as a round number. The published data does not describe a mix. It describes a ranking with gaps in it large enough to make the mixing decision almost irrelevant. Acquiring an email address costs $2.81 on average, $3.64 on Meta and $86 on display — a thirty-fold spread inside one table. Return on ad spend runs $2.48 on search, $1.82 multi-channel, $1.11 on display and $0.17 on Google Grants. Against that, the sector reinvests ten cents in digital advertising for every dollar of online revenue, so most organisations are spending little enough that spreading it across four channels guarantees none of them reaches a size where it can be judged.

An email address costs $2.81 on average and $86 on display. That is not a channel preference, it is two different activities sharing a row in the same table.

1. 01

   The spread inside the table is wider than any optimisation you could run inside a channel

   Cutting display's $86 acquisition cost by a third still leaves it twenty times the average. When the differences between options are that large, sequencing beats tuning: fund the top of the ranking to a size where it can be measured before funding anything below it.

2. 02

   Google Grants returning $0.17 per dollar is not a failure, it is a category error

   The Grant is free inventory with restrictions attached, and measuring it on return on ad spend compares it to money the organisation actually spent. It earns its place as a discovery and advocacy channel with its own success measure, and it should never be averaged into a paid return figure that then guides a real budget decision.

3. 03

   Email carries 11% of online revenue and is the destination the ads are buying

   Revenue per thousand fundraising emails is $54, revenue per subscriber is $2.40 a year, and email is 11% of online revenue at fifty messages per subscriber per year. Acquisition advertising in this sector is mostly list building, so its return lands in the email programme rather than in the campaign it was attributed to. An organisation judging acquisition ads on immediate gifts will switch off the thing feeding the channel that does the work.

4. 04

   Advocacy draws four times the click-through of fundraising and is the cheapest front door

   Advocacy email runs 2.3% click-through and a 1.4% response rate against 0.59% and 0.05% on fundraising email, with 0.34% of advocacy actions going on to a donation. As an acquisition route that is strong and slow, and it has to be measured as a two-step path or it will look like it does not work.

5. 05

   At ten cents reinvested per revenue dollar, the constraint is minimum viable channel size

   A programme spending at the sector reinvestment rate cannot fund four channels to a level where any of them produces a readable result. The model sets a minimum spend per channel below which the channel is not run at all, rather than trimmed.

Cost per email acquired: **$2.81** overall, **$86** on display · return per $1: **$2.48** on search, **$0.17** on Google Grants.

The model funds the ranking in order instead of allocating a share to each channel. Phase one sets one definition of an acquired email address and one attribution window, then measures the organisation's existing channels against it — including the Grant, reported on its own terms rather than as paid media. Phase two funds search and the best-performing acquisition channel to a minimum viable size, defined in advance as the spend below which a result cannot be read, and pauses anything that cannot be funded to that level. Phase three connects acquisition to the email programme so the return is measured where it lands: revenue per subscriber over twelve months, rather than gifts attributed to the ad that acquired them. Advocacy runs as a deliberate two-step entry path with its own reporting.

1. 1

   Week 1

   Definitions and attribution window

   One definition of an acquired address, one window, applied to every channel

   Owner: OmniFlow + organisation

2. 2

   Weeks 1–4

   Channel baseline

   Existing channels measured against the definition; the Grant reported separately

   Owner: OmniFlow

3. 3

   Week 4

   Minimum viable size

   A written floor per channel; channels below it are paused rather than trimmed

   Owner: OmniFlow + organisation

4. 4

   Weeks 5–20

   Fund in ranking order

   Search first, then the next channel that clears the floor

   Owner: OmniFlow

5. 5

   Weeks 8–24

   Connect to the email programme

   Acquisition cohorts tracked through to revenue per subscriber over twelve months

   Owner: OmniFlow

6. 6

   Month 12

   Review

   Cost per acquired address and twelve-month revenue per subscriber, by channel

   Owner: OmniFlow + organisation

The model reports cost per acquired email address by channel and twelve-month revenue per subscriber by acquisition cohort, and it reports the Google Grant separately with its own measure. It does not report a blended return on ad spend, because blending free inventory with paid inventory produces a number that cannot guide a decision. Channels that cannot be funded above the written floor are named in the report as paused rather than quietly underfunded. Published figures are replaced by the organisation's own from quarter two. Modelled outputs are not a forecast or a guarantee of results.

Modelled. Inputs: M+R, Benchmarks 2026 (2025 data). Modelled outputs are not a forecast or a guarantee of results. Cited: M+R *Benchmarks 2026* (200+ nonprofits).

Every figure below is published research, not a client result. They are the inputs to the arithmetic above, listed so it can be checked rather than taken on trust. The bracketed number points to the full citation at the end of this page.

These are the surfaces this engagement is run and measured from, shown with representative figures built around the benchmarks cited on this page. Every account we run reports into views like these, and you keep ownership of all of them.

Each figure on this page, the system it is read from, and the definition and window it is measured over.

Not applicable — this is a modelled engagement. Its weakest input is that the published advertising figures are dominated by large national organisations with brand recognition a local or regional organisation does not have, and display in particular is often bought by those organisations for reach rather than for list building, which is part of why its acquisition cost looks the way it does. The model uses the ranking to set an order of operations, not to predict what any individual organisation pays.

A modelled capability, not a client account. Figures illustrate what the model produces and are labelled as modelled wherever they appear. See all our work

Full citations for everything cited on this page, with the sample and period each study covers, so you can go and read the original.