5 Mistakes You Could Be Making With Your Facebook Ads

1. Not Enough Creative Testing

Not doing enough (or any) creative testing means you’re leaving money on the table for your business.

Creative testing is one of the most impactful ways to improve performance, as even marginal lifts in performance can mean big returns in the long-run.  

Our internal data across Consumer Brands estimate that effective volume of testing accounts for a minimum 23% increase in KPI.

It requires testing different elements within your ads, like the headline or the hook, to improve metrics like click-through rate and conversion rate. Ads with strong click-through rates and conversion rates ultimately create better returns on investment.

When you only launch one version of an ad, you increase the likelihood that the ad will fail. The reality is that sometimes the winners of the test are the ones you would least expect… the point is that even the best creative teams don’t pick the winner 100% of the time. By testing multiple versions of the creative, you are hedging your bets on the test.

Here’s a recent creative test we’ve run:

‍
Results: Version C drove a 28% higher CTR and a 24% lower cost per purchase than the other versions.

‍

2. Not Optimizing for the Correct Campaign Objective or Conversion Event

Meta has the most powerful algorithm in advertising, but it can only work effectively if we are giving it the correct inputs.

Meta has recently narrowed their campaign objective options from 11 different objectives down to 6, which are really asking what business goal you are trying to achieve. Don’t select Traffic if you are really wanting to drive purchases, as you’re really just going to get a ton of unqualified users to your website.

The Meta algorithm is like a dog with a bone. If you tell it to drive traffic it will drive traffic without concern for purchases or other objectives.

The campaign objectives are pretty self-explanatory, but it is important that once selecting your campaign objective, you then determine the correct conversion event at the ad set level.

In the majority of cases, if you were to select Sales as your campaign objective, you don’t want to select Add-To-Cart or Initiate Checkout as your conversion event. Telling the algorithm to optimize for an event higher up in the buying journey when you’re really trying to get purchases will result in the algo spinning its wheels. If you want purchases, optimize for purchases.

Important to note that the ad platform requires about 50 conversion events to be able to out of the learning phase and really optimize performance. Meta’s guide to the learning phase.

‍

3. Only Using the Default Attribution Model

Selecting the correct attribution model for your business can make or break your campaign performance.

By default, Facebook selects a 7-day click or 1-day view attribution model, which means that it will count a conversion if a user has clicked the ad within the last 7 days or views it within 1 day. While this can work for some businesses, others, like most of those in the ecommerce space, should be testing the other two attribution model options: 7-day click and 1-day click.

We’ve found that when using the 7-day click/ 1-day view model, the algorithm heavily relies on those who view and convert, which can lead to less incremental conversions. When pivoting to a click-only based model, the algorithm actually changes how it optimizes and tends to work better for prospecting new customer acquisition.

In one study we completed, the 7-day click campaigns produced a click-through ROAS 430% higher than 7-day click/1-day view.

‍

4. Focus on Extremely Niche Audiences

Audience strategy has done a complete 180 from a couple of years ago when testing a multitude of niche audiences was the name of the game. Post launch of iOS14, Facebook has much more limited tracking for app and web events, reducing the quality of these small audiences.

When audience sizes are small, not only are your CPMs going to be higher, increasing your top-of-funnel costs, but that audience has a limited lifespan, which is not suitable for long-term, stable performance. You are handicapping the algorithm by limiting what audience you can reach.

So how do you make sure that you’re serving ads to the right people? To revisit the impact of iOS14, tracking off-platform activity is now limited, which means that the in-app user behavior, like ad engagement, is now much more important to performance. This gives Facebook the upper-hand in knowing who will be the most likely to convert on the ad and can match users to the right ads much more efficiently than we can by testing niche audiences.

Increasing your audience sizes or even going fully broad (outside of geo, age, and gender adjustments)lets Facebook analyze the context of your ad creative to do the targeting for you.

‍

5. Only Using PDPs or PLPs as Your Destination

Even if your ad account is fully tuned up, sending potential new customers to a Product Detail Page (PDP) or Product Listing Page(PLP) can be a jarring user experience. This is particularly important if you have a product or service that requires further education or the value proposition needs to be further expressed.

Landing pages can be a great option to build out user journeys that are prime for conversion. For example, does your customer need help deciding which of your products would be best for them? Rather than increasing friction by forcing the customer to do the research for themselves on your website, try a quiz that personalizes the results to that individual. Is your product in a highly competitive vertical and require more trust in order to get customers to ditch their current brand for you? Try a Listicle landing page that builds social proof and explains your value proposition.

Here’s a recent landing page test that we’ve run:

We found that option 2, the Listicle, drove a 51% lift in conversion rate for new customers.