SPN. 137: How to Improve Fund Raising Performance

Plus, pMax comes of age; becoming an early adopter; watching trends not exact numbers

Jan 27, 2025

A very warm welcome to all the new subscribers.

You’ve joined a community of 2k+ marketing and fund raising operators at mission-driven Org’s. I’m thrilled to have you as readers and truly appreciate your feedback and support.

🚀 Big News in Nonprofit Tech! 🚀

Huge congratulations to Fundraise Up on securing a $70M growth investment 🎉 Nonprofits deserve world-class fundraising technology and Fundraise Up is a brilliant example.

As a very early customer and cheerleader for more than 5 years, it’s been inspiring to watch how their first-class people and first-class platform have transformed online giving for nonprofits (and donors). I could not be more delighted for them.

Having a technology partner that’s innovative, growing, and committed to driving impact is rare as much as it is transformative. Fundraise Up is that Partner.

Get a conversation with them on the books!

Game-changer? Every. Day!

In this week’s SPN →

Reaching audiences in an Ad Tech fragmented world
5 things you should know about PMax
Approaches I deploy today to improve fund raising performance
Measurement – look at trends, not exact numbers

and, plenty of Jobs & Opps that took my fancy this week.

Let’s jump in!

5 Gems: Performance Max

In SPN #29 I tried to demystify pMax a little and make sense of the black box that it is - worth a read as a refresher (there’s also a relevant piece in there about taxonomy and why it matters, and the value add off agencies).

This EOY, pMax went gang busters for me (at least I think it did)*. I’ve spent this month looking at a lot of EOY data and insights, and seen in other accounts a lot of fundamental mistakes being made.

So here are 5 gems teams must know about pMax (that might sound obvious, but most people seem still unaware of):

You can add negative keyword (lists) via Google Support.
Audience signals are NOT direct targeting.
Search themes can lead to cannibalization of Search.
pMax over-inflates itself with brand conversions.
You can see spend allocation across channels with a script.

Let’s jump into each of these!

1. You can add negative keyword (lists) via Support

Since the beginning, we haven’t been able to add negative keywords to pMax campaigns directly.

However, there’s a workaround: you can fill out Google’s Performance Max Campaign Modification Request Form to get their Support team to add negative keywords or negative keywords lists to your pMax campaigns.

= a great way to exclude brand traffic or underperforming search terms.

It’s a simple form that you can fill out to ask Google to exclude any keyword, topic, or placement.

It speaks for itself:

Fill out all relevant details about your account.
Select what you want to exclude and at what level.
Download and fill out the “pMAX: Request Template”
Upload the template, give consent and submit the form.

Ta-da, Google Support will now add your negative keywords (lists) to your pMax campaigns!

p.s. Self-serve campaign-level negative keywords for Performance Max are available in Beta right now. Hopefully they’ll soon be rolled out to all accounts.

2. Audience signals are NOT direct targeting

With pMax campaigns, you can set up Audience Signals to inform Google what audience segments are valuable to your Org. However, take note: audience signals are NOT direct targeting.

They’re mere signals, used to guide Google's algorithm in the right direction. Google can and WILL go beyond audience signals.

In fact the official documentation confirms it: “Performance Max may show ads to relevant audiences outside of your signals if they have a strong likelihood of converting to help you meet your performance goals.”

It surprises me how often I still hear people talk about complex audience signal layering strategies for pMax, as if it has any impact on campaigns. It doesn’t and it's a waste of time. Audience signals are just suggestions that give you a fake sense of control.

They’re NOT used for targeting, and Google WILL eventually go beyond them.

3. Search themes can lead to cannibalization of Search

When you run Search and pMax campaigns together, this happens a lot:

Search stops spending
pMax takes over most traffic

And it can be frustrating, especially when you want to control your traffic with well-structured Search campaigns. Understanding the ins and outs of pMax, its mechanics, and how internal matching works between campaign types is a whole to-do.

But in short when you use Search Themes in pMax, a seemingly harmless feature, you basically tell Google that you don’t mind whether a specific search term gets picked up by your Search or pMax campaigns (because they get the same prioritization in the auction as your identical Phrase and Broad Match keywords).

This graph from WARC is the most helpful I’ve found in explaining how the Internal Matching Hierarchy works between Search and pMax campaigns.

**4. *Performance Max over-inflates itself with brand conversions**

You can’t trust your pMax results.

Why? Because they’re heavily inflated by Branded Search conversions, which makes pMax look better than it actually is.

pMax heavily cannibalizes your other campaigns and LOVES to take credit for every single conversion. People who search for your Org would’ve converted anyway, so why should you let pMax take credit for it?

I think you should exclude branded traffic, but if you’re agency or internal team think differently and brand is still included in your pMax campaigns here’s a simple solution:

Analyze your share of brand in pMax (with a script)
Exclude brand from pMax with Brand Exclusions or via Google Support
Catch brand traffic in Branded Search and Branded Shopping campaigns

5. You can see spend allocation across channels with a script

pMax is a giant black box and Google isn’t telling us how much we spent on each individual ad network (Search, Shopping, Video, Display, etc.).

This script available on Github allows you to see the spend allocation and performance per network. It’s an absolute MUST in every account that runs pMax campaigns and shows you:

Where Google is showing your ads (shopping, display, video or search/other)
Which search categories your ads are showing for (sorta like search queries)
Account-level insights with metrics for each of the four channels
Dive deeper and see the channel split over time

And that’s your 5! Hopefully a good nugget in there to test into and implement.

Jobs & Opps 🛠️

Marie Curie: Senior Digital Marketing Manager (£44,550 - £49,489 (+ £3,500 London Weighting))

USA for UNHCR: Director, Audience-Centric Engagement ($169,612 - $185,031)

UNICEF UK: Senior Innovation Manager $169,612 - $185,031

Diabetes UK: Senior Community Fundraising Manager (£47,348 - £52,609)

Greater Ormond Street Hospital: Head of Mass Participation (£62,519)

American Diabetes Association: National Manager, Peer to Peer

GoFundMe: Senior Social & Community Manager ($88,000 - $132,500)

Fundraise Up: Director of Fundraising Strategy & Success ($130,000 - $200,000)

UK for UNHCR: Board of Trustees

World Food Programme: Internship - Technical Project, Individual Giving Team, Private Partnerships Division (Monthly stipend)

Marlene Meyerson JCC Manhattan (MMJCCM): Executive Vice President ($250,000 - $300,000)

→ More jobs updated daily to SPN’s sister website: www.pledgr.com

Reaching Audiences in a Fragmented World

Most Org’s tend to cluster into the same handful of channels - our “safe zones” - Meta Ads, Google Grant campaigns, and Google Paid Search / GDA.

Small and large Org’s spend 95% of their budget across these three channels. But for at least 2 reasons these platforms aren’t enough to hit digital fundraising performance goals any longer:

Costs will continue to rise. There is no reason for the price of media on any of the major platforms to go down any time soon, with Big Tech revenues consistently trending up yearly. Advertising competition is getting fiercer. Combined, this hurts the unit economics of digital fundraising and puts increased pressure on already not-so-good CPAs.
AdTech is getting more fragmented. “Google Display Network” – now GDA – used to be enough to reach 90% of US website users and command their attention across most of their daily online activities.
1. Now? Not so much, with the consecutive rise of Meta, TheTradeDesk, Amazon, TikTok, and whoever will be the next player to chip away at the market- and attention-share.
2. Antitrust cases and privacy laws don’t help; further breaking down (for the better!) the carefully monopolized digital marketing landscape by deprecating 3^rd party cookies and forcing publishers to step up their advertising game individually, instead of relying on selling inventory to programmatic ad exchanges in bulk.

So while the “advertising buying” market is significantly consolidated, the “advertising consuming” market has become highly fragmented.

The donor journey is getting longer and involves more touch points (SPN #130). Orgs that limit themselves to “the mighty duo” of Google and Meta are paying monopoly prices for, at best, a market challenger reach.

This is still a good deal if your Org is only launching a digital fundraising program now. There’s still money to be raised in the Google / Meta universe. With little understanding of target audiences and no historical data for targeting, Google’s / Meta’s AI algorithms, pMax and Advantage+, are a safe bet (scroll up).

But if your Org’s fundraising program is more mature, here are three approaches I’ve deployed to improve my fund raising performance, which should stand you in good stead in 2025’s fragmented market.

Measurement – look at trends, not exact numbers.
Become an early adopter of new platforms and use more platforms.
Prioritize brand-building over deploying the same targeting.

Approach 1

Chances are your Org is using Google Analytics 4 as a web analytics platform of record.

The issue there is GA4 uses modeled data for User count. To make matters worse, the exact percentage of modeled versus deterministic data for your particular dataset is unknown. And it’s changing over time as Google moves away from cookie-based measurement.

There’s no way to look under the hood and know exactly how Google is modeling the data to stitch it together. The donation count your Org sees in the CRM or the donation platform is likely at least 5% different than that same number in GA4.

The above scenario isn’t good or bad – it’s a reality given that the concept of “User” is fundamentally flawed in digital advertising. I see too many Org’s spending too much time trying to match the data across all the platforms, which just paralyzes any decision-making process… until it gets within some magical 2% discrepancy (not sure why Org’s choose 2% but seems to be a thing!).

If the above resonates, stop wasting time. Freeze the implementation and look at the “users” your fund raising channels drive to the website and the number of donations attributed to those channels as trends instead of exact numbers. As long as both numbers increase faster than the budget, it works as it should.

SPN Tip: Consider using “sessions” in GA4 instead of “Users” and pulling it via API instead of UI. This is the closest you can get to unsampled data in Universal Analytics. You’ll miss out on most of Google’s abilities to stitch sessions across different devices within the same household but this approach will give you a different lens.

Approach 2

Take Threads by Meta, for instance: your Org might have a smaller following there than on X or Facebook / Instagram proper.

But that small audience is likely an audience of your Org’s core “brand loyalists” who want real-time updates (organic social management) and will be the best-performing base for a look-a-like acquisition campaign (paid social).

Test reusing content you’ve already created for other platforms – just recut your IG reels for TikTok and YouTube Shorts.
Post on Threads, BlueSky, and X all at the same time.
Launch an ad on Amazon DSP (targeting only Amazon.com inventory) alongside every GDA ad.

In a world where creative is targeting (SPN #136), posting – and promoting – the same content across these different but mutually exclusive platforms is going to allow your Org to create fuller surround sound around prospective donors across all of them.

SPN Tip: with TikTok’s currently unpredictable fate, Facebook, Instagram, and Snapchat are prone to benefit in the short term and gain a higher share of attention. If your Org is absent from some of them, e.g. SNAP, now is the time. Check out SPN #102 for what works well on SNAP.

And if the cause your Org serves doesn’t conflict with Red Note’s policies, check it out as well.

Approach 3

Each platform has its own set of targeting dimensions, and new platforms tend to be less developed.

Netflix barely has look-a-like capabilities, and Spotify is only starting to build its advertising suite. They can’t compete in targeting accuracy with Google or Meta platforms, and yet they command a massive share of your donors’ attention every day.

Test new tools available today, instead of waiting for them to improve.

Allocate a certain % of your budget (time and money) every month to tinker with the less-tested and explored platforms. I like it to be 15% of my budget.
Split it across at least two platforms.
Pick a few mutually exclusive markets to test these platforms in. For example, launch one in the Houston metro and one in Austin. It’ll resemble a (simplified) version of a proper incrementality test.
Try to target substantially similar audiences between those two. For example, use “show genre” on Netflix and “podcast category” on Spotify to reach users consuming similar content.
Compare user growth trends and downstream metrics in those markets in GA4 after a month of running campaigns.
Pick the winner – or keep running both if the increase is comparable and substantial enough.

As a rule of thumb, I’m looking for 0.3:1 ROI in the first month (SPN #130) that should then become 0.5:1 in Month 2, 0.75:1 in Month 3 and break even in Month 4.