Traffic isn't targeting: A nonprofit guide to Google Ad Grants

Most nonprofits still run Google Ad Grants the same way they did five years ago: pick keywords, hope donors search for them, and wait for clicks. That looks like a digital strategy. It isn't. It leaves money on the table. That was the starting point for our webinar this week with Jason King, Certified Google Ad Grants Trainer and Salvatore Salpietro, Chief Growth Officer at Dataro. Here's what stuck with us.

Three ways the old playbook stalls

The same three assumptions show up again and again in nonprofit ad accounts, and they all rely on waiting for a donor to make the first move.

• The keyword trap: You display a sign to fish that says "worms for sale." It reaches people already looking for what you offer. That intent is valuable, but it is strictly limited to existing demand.

• The lookalike limit: Most fundraising teams build lookalikes from past behaviour. They export last year's donors, or everyone who gave more than $50, then ask Google or Meta to find matches. It mirrors your existing base. You get incremental lift, not scale.

• The site-speed blindspot: Teams launch great ads but leave the website slow. A laggy landing page triggers temporary suspensions and hurts organic search ranking. Compressing large images is often the quickest fix, but it gets ignored in favour of tweaking ad copy.

None of these blockers are about the grant budget itself. They are strategic. So what does fixing it actually look like?

Performance is a system, not a switch

Good ad performance runs on three connected parts: miss one, and the other two stop delivering.

• Keywords harvest demand that already exists. They capture the low-hanging fruit from people who already know your cause or your exact niche.

• Audiences create demand. Instead of waiting for intent, you put your cause in front of people whose profile and behaviour suggest they would respond. You are identifying the kind of person likely to give, even if they weren't searching for you.

• Propensity points to what is likely to happen next. Instead of asking who gave before, it draws on billions of behavioural signals to predict who is likely to give next. In charities, where 20% of donors typically drive 80% of revenue, predictive models give ad networks an incredibly rich signal to find your next supporter.

Miss keywords and you lose easy wins. Skip audiences and you hit a hard ceiling on growth. Skip propensity and you waste budget on generic lookalikes. All three, moving together, is what makes ad spend work.

What an ad account needs before it earns the right to scale

An ad strategy is worth scaling once it clears three bars:

• The right campaign type: Run both search and Performance Max. Traditional search campaigns force you to guess keywords. Performance Max drops keywords entirely and builds a profile of your audience. Jason noted that PMax now carries about 75% of the average grant budget, up from 66% just a few months ago.

• First-party data signals: Feed Performance Max your own data. Uploading a CRM list or a CSV of at least 100 active contacts lets Google build a lookalike audience from a position of strength. Note that grant accounts cannot do true remarketing; they target people similar to your list, not the list itself.

• The landing page experience: Look at literacy nonprofit Room to Read as a model. A donation page needs a clear match offer at the top, trust signals, a direct ask, suggested gift amounts tied to real outcomes, and emotional testimonials. If the landing page doesn't convert, don't scale the ads yet.

Clearing those three bars is one challenge. Deciding how to balance your free grant budget against a paid ad budget is another.

Where does paid ad spend fit?

One of the most important takeaways from the session was exactly that: what should stay within the free grant, and where does paid budget fit? The honest answer is they are not rivals; you need both. The grant provides $329 a day of in-kind search advertising, but grant ads sit below paid ads, and you cannot run true remarketing. Use the grant for what it is best at: building awareness, testing hooks, and capturing audience lists. Save your paid budget for running true remarketing campaigns against those lists. That is where donations scale.

What this looked like in practice

Want to see it in practice? Watch the full recording to hear Jason King share the December results from Room to Read to see the direct contrast between the two accounts. The grant spent nearly $8,660 and raised about $4,480. The paid account spent just under $29,000 and raised a staggering $368,000 by leveraging true remarketing against the organization's own supporter data.

Watch the on-demand replay