Can I Really Just “Go Broad” On Facebook?

“Going broad” is a hot topic in Facebook Ads land, especially after iOS 14. For some brands, interest targeting and pixel-based audiences just don’t seem to work anymore. So is the solution to simply abandon audience targeting altogether?

I might ruffle some feathers with this one, but I think the answer is “it depends”, not a resounding “HELL YEAH!” To explain my position, I’m going to take a step back and explain how Facebook finds your audience and how “go broad” works within that framework.

How Facebook Ads Work

A few weeks ago I posted a thread on Twitter outlining the “dartboard analogy” of how Facebook’s advertising product works. It was pretty popular, so I’ll rehash it in a bit more detail here.

You can visualize Facebook’s user base as a dartboard. But instead of rings around the target, this dartboard contains millions of tiny dots that are constantly changing position on the board. Darts that represent users with similar demographics, psychographics and interests tend to stick close together, although that’s not always the case.

You pay Facebook for the chance to throw a dart at the board. If you hit the right dot, you make a sale. But from where you stand, you can’t see any of the dots on the board because they’re too small. It just looks like a big white circle to you.

You stage the conditions around the throw and pass that information to Facebook. But they’re close enough to the board to see the darts, so they make the actual throw. Neither you nor Facebook know exactly which type of dots you need to hit. But, because Facebook knows that similar dots cluster together, it can refine your aim after multiple throws.

When you pay Facebook for the chance to throw a dart, you’re also competing against a bunch of other people for a limited total number of darts. That’s where the auction comes in; the more people want to throw (advertise), the higher the price of a single throw.

So, when you are able to run ads profitably, a few conditions are lining up perfectly:

1. You’re giving Facebook enough information to line up the throw correctly
2. Facebook has a decent view of the dots
3. The price of a throw is reasonable
4. There are enough dots on the board that would ever convert on your product and offer

The fewer throws it takes you to hit a dot, the more profitable your ad spend. It’s a very dynamic situation. So dynamic that it’s amazing that Facebook ads work so well for so many businesses. 

The privacy measures that were instituted with Apple’s iOS14 software update did two things:

1. Partially obscured Facebook’s view of the dartboard. It became harder for Facebook to determine who might be a good target for a given dart thrower.
2. Made the dots harder to hit. There was less total information available about a given Facebook user as cross-device tracking became more difficult.

Learn how Triple Whale makes Facebook targeting more effective for ecommerce brands.

How Does “Go Broad” Factor Into This?

When advertisers are advised to “go broad”, they’re being told to run ads with as few targeting settings as possible. Sometimes this means all US Facebook users over 18. Now that’s broad!

What is the rationale behind this? If Facebook has less information about each user, and the consolidation of user behavior across individual cookie IDs is less effective, then activity that happens within Facebook’s own “walled garden” is the best way to match a user to the right ads. 

To put it simply: Facebook will use a viewer’s engagement with an ad to find the right audience for the ad. If someone clicks an ad and/or converts on it, Facebook will show the ad to more users like that.

To revisit the dartboard metaphor: when you use interest targeting, you tell Facebook to start aiming at a specific “slice” of the dartboard, because there’s a higher probability your relevant audience resides there. When you use broad, you tell Facebook to split the dartboard in half (then a quarter, then an eighth, etc.) until it hits a dot, then start aiming around that dot.

In the broad targeting model, your ad creative does most of the heavily lifting. The more your ad creative resonates with a specific audience segment, the faster Facebook hits your first dot.

When you’re encouraged to launch ads with broad targeting, the subtext is that Facebook will find the right audience for you faster starting from scratch than it will with any sort of assistance. And that is because Facebook’s information on its own audience has become so muddled that starting off with assistance may actually be sending the algorithm on a wild goose chase.

Facebook’s own “best practices” advocate for a broad approach. [link 2]: 

• Give Facebook as many assets as you can
• Let Facebook choose the placements
• ???
• Profit!? 

The ??? is Facebook’s black box algorithm, which is designed to match the right creative with the right users.

Can All Brands Go Broad?

I’m always wary of any “best practices” that are framed as the answer for every brand, no matter the growth objective or financial circumstances. In my opinion, not every brand should go broad. There are some brands that can’t go broad profitably, and there are other brands that may benefit more from a lower spend, higher return approach.

If you’ve ever watched Shark Tank, you’ve probably seen Lori Greiner [link 3] give contestants two types of feedback: “This product is perfect for QVC!” or “This product will never work on QVC, so I’m out.” 

Lori has unlocked the magic of product-channel fit, and that’s why she is a millionaire (lesson in there…). Some products will naturally resonate more with the people who watch QVC regularly. And some products lend themselves better to QVC’s format. Facebook has the same nuances. Some products will always convert better with Facebook and Instagram’s most engaged audiences (women over 35). 

Some products have such a small audience that “going broad” becomes prohibitively expensive. Facebook has to divide the dartboard a lot of times to find that first hit, jacking up your cost per conversion. And then it turns out that there aren’t a ton of relevant dots near the first hit, so the algorithm needs to start from scratch.

The more expensive your product is relative to its competitors, and the more specific the use case for the product, the more issues you can anticipate with “going broad”. Some examples of products that probably wouldn’t be able to “go broad” successfully: a high quality beekeeper’s helmet, a $10,000 crocodile skin handbag.

In other cases, “going broad” may be profitable, but it won’t be the most efficient strategy a brand can pursue at its desired scale. Not every brand is able to spend a million dollars a month on Facebook ads. If your total monthly budget is limited by cash flow and inventory, a more targeted approach could give you a higher return than broad–at that spend level.

If your brand has a diversified marketing mix, a sizable wholesale or physical retail presence, or a large base of loyal customers, you might not need to max out acquisition every month. These brands can afford to use Facebook as a supplement to their customer acquisition efforts at a lower monthly spend, and broad might not be the most efficient way to do this.

Finally, there are some scenarios where you want to create a funnel that captures customers with high lifetime value potential. Creative targeting can achieve this to a certain extent, but audience targeting has the potential to improve efficiency. If you run dynamic ads for broad audiences, Facebook will typically prioritize your “lowest hanging fruit”: low AUR products sold to impulse shoppers with low lifetime value potential.

Learn how Triple Whale makes Facebook targeting more effective for ecommerce brands.

Targeting After iOS14

If your business is at least six months old, and you have a total customer base of at least 5,000 people, it’s worth testing Lookalike audiences modeled from an upload of your customers’ emails. Ideally, the seed list you upload to Ads Manager contains at least 2,000 people. If you have enough customers to segment out your best shoppers–those who have returned multiple times and purchased mostly at full price–all the better.

This strategy won’t necessarily allow you to scale your ad spend to seven figures a month. But it does have the potential to bring in a semi-predictable stream of new customers at an affordable cost–if your ad creative and landing pages are on point. And after the disruption of iOS14, that’s exactly what many brands need most right now.

I know some of you reading this are all-in on broad and are about to pelt me with a bucket of rotten tomatoes. I’m not saying this strategy will work for everyone, but it’s worth adding to your list of experiments if nothing else seems to be working.

This does present a challenge to brands that are starting from scratch. I have two pieces of advice here. The first: if you’re starting a brand and plan to use Facebook as your primary customer acquisition channel, pick a product category and price point that “works” with the broad approach. Don’t try to launch your avant-garde couture brand using Facebook ads.

The second: if you’re absolutely dying to launch a product that sits outside of Facebook’s sweet spot, try to win your first 100 customers elsewhere. Or write your business plan with the understanding that Facebook will not be a profitable acquisition channel for your brand. 

The days where you could sell anything profitably on Facebook with a few simple targeting hacks are long gone. But that doesn’t mean that broad is the only option, or the right option for your brand at this moment in time.