4 Deadly Mistakes To Avoid When Running Ads on Meta, Tiktok & Google

Here’s what they don’t tell the average advertiser: most advertisers who advertise online fail at achieving or long-term maintaining their advertising objective, which is to grow their business through profitable paid ads. There are many reasons for this, which are primarily controlled by the market of supply and demand. However, investing your money on the wrong advertising platform that you think is driving you the best results instead of advertising in places that actually drive real results can also contribute to this failure.

Aside from being in the right place at the right time and choosing the most efficient advertising platform that can create the best return on every dollar spent on ads, there are also other things that you should be aware of that could dramatically improve the performance of your ads, regardless of the ad network. Even if you are investing in the most effective ad platform for your product, here are some deadly mistakes to avoid when advertising across one or more of the advertising platforms. This post reviews the three biggest platforms to avoid an overly lengthy article, but the same type of rules apply to smaller advertising platforms as well, which for the most part use the same type of logic and advertising interface.

Mistake 1: Running Ads with View-Through Conversions

If you have read my previous posts, you most likely understand how I feel about the use of view-through conversions, which I define as a way for advertising platforms to take credit for conversions (results) that they never delivered or were directly responsible for. To really understand why this can do more harm than good, I recommend that you view one of my dedicated articles about view-through conversions, but to make a long argument short, just think of view-through conversions as getting credit for scoring a point in a sports game without actually being the one kicking or making the final shot touch but still getting the credit for it.

If you are advertising on more than one platform and both are counting view-through conversions, both players could get credit for the same point even if there is only one. It naturally gives more points (results) to the ad platform than it should because it doesn’t just count the number of times people clicked on an ad and completed a purchase or a conversion; it also counts the number of times they viewed the ad and then completed that action. This is highly manipulative. Aside from causing overly inflated results in terms of comparison to the actual results, it also trains the ad network’s algorithm to invest more in conversions that are likely to generate more view-through conversions compared to click-through conversions because these conversions could be easier to get. Instead of investing all of the ad impressions that you are paying for in ad impressions that could generate a conversion after someone actually interacted with that ad and clicked on it, it’s essentially training the algorithm to maximize results for the ad network as a first priority. If you are a sophisticated computer, you can predict when users see a product and when they are likely to come back and complete the purchase. You just have to make sure that an ad impression appears between that time gap to be able to take credit and say, "Hey, I drove that conversion; it’s because of me," when in reality a user might not have noticed the ad or might have already completed the purchase regardless.

View-through conversions might not seem like a huge part of your ad budget investment or the number of conversions that they generate, but just estimate how much ad impressions and budget they will eat up over time and how much better off you would be if you invested all that budget in ads that are likely to generate a conversion as a result of interacting with them, not just placed in a way that has a low likelihood of impacting the intent of the user and has little to no impact on the outcome of the user after seeing the ad.

Mistake 2: Enabling Partner Networks as a Placement

Advertising networks are known to work in sneaky ways that might seem like they are helping advertisers reach their goals more efficiently. If you are not yet familiar, partner networks are the networks that big ad platforms partner with to deliver ads outside their own network and on the network of their partners.

On Meta, it’s called the Audience Network.

On Tiktok, it’s called the Pangle Network.

On Google, it’s called the Search Partner Network.

Ad networks rely on partner networks for a number of reasons that mostly benefit them more than they benefit their advertisers. The first reason is obvious: it’s a way for them to generate more revenue outside of their own platforms by extending their ads to third-party websites. It also gives them more leverage as they help third-party websites make money and monetize their website and content by making money without having to do anything but place a line of code on their website and start making money. All these ads that you see on free apps or browsing through websites are app partner apps.

Google has the biggest ads network, but Tiktok and Meta are growing their own rapidly as well. The second reason that they use ad partners is to have more touchpoints between users and the advertiser that is placing ads on their ad networks. If I am a shoe brand and I am advertising on Google, for example, aside from YouTube and Google Search and Gmail, Google is pretty limited when it comes to creating another touchpoint interaction with that user. If it’s trying to reach that user with an ad because it predicts that this user is or might be in the process of purchasing, it can position the ad in an app that a user is using on their phone right before that purchase is made. Ads on mobile are extremely annoying, and the first thing that you try to do when you see them is click the X button, but at the same time, many times it results in you accidentally clicking on the ad. Clicking on the ad will attribute a conversion if you end up purchasing that product in the following days regardless of that app impression in the mobile app.

It extends the ad network’s ability to take credit for the conversion and say, "We drove it," but also dilutes the efficiency of the ad. If you ask a room full of advertisers how many of them are advertising on platforms like Google and want to show their ads on apps and partner networks, I think the majority of them will say that they had no idea their ads are running outside of Google assets. And it’s no different for advertising on Tiktok & Meta; most advertisers wouldn’t knowingly choose these additional placements. On Google, it’s actually mandatory to include these placements under some of the ad formats like YouTube video ads and Performance Max campaigns. To exclude this, you actually have to do a little heavy lifting and manually exclude these app partners under the exclusions placement list for each individual campaign.

On Tiktok & Meta, it’s optional, but it’s also the recommended default option, which is to include every possible placement. Another reason that ad platforms use it is to lower the average cost per impression that advertisers pay and make it seem like the cost per impression is lower. What advertisers normally don’t consider is that the cost per every 1,000 ad impressions they pay is made up of a summary of lower-cost ads on the partner networks and higher costs on the actual ad platform that they are advertising on. So ad platforms use audience networks to provide extended and cheaper reach. Should you exclude them? Absolutely. When you sign up to advertise on Meta, you want your ads appearing on Facebook and Instagram, not on some shady-looking random websites.

Will you miss opportunities by excluding them? If you look at the stats, in some cases it might appear that you are receiving some results from those partner websites, but in reality, much of these results are artificial and your money is better invested when it’s invested directly on the ad platform that you are advertising on. While on Tiktok & Meta, it’s easy to remove and manually select placements on some formats. On Google, like Video or Performance Max, you are forced to include Google’s partner network. Unfortunately, excluding is a little harder, and you actually have to go to placements exclusions and exclude all the mobile apps categories and everything else that seems like it is not YouTube or Google.

Mistake 3: Using Different Attribution Windows

I wrote a specific article that goes into length and emphasizes the importance of using the same attribution models across all platforms. But it’s important to mention again here. Attribution models are how ad platforms like Facebook, Tiktok, and Google measure their results and optimize their ads. They define what and when counts as a conversion (e.g., 7 days after a click or 1 day after a view), and that’s how they report results to you.

What’s the problem here? If one platform attributes based on conversions that happened within 7 days after a user clicked on an ad and another ad platform attributes after 30 days after a user clicked on an ad, how would you define the value of each one of those conversions? Are they worth the same? No, they don’t. Like the first point I mentioned, view-through conversions are worth very little, but if someone clicked on an ad and purchased the next day vs. someone who clicked on an ad and purchased 20 days later, do both of these conversions have the same value? No, they don’t. That’s why it’s important to define all attribution models in the same settings. Whatever the most effective setting is for your business and industry, make sure all your ad platforms are set on the same attribution model.

Here is how it looks on Meta:

Here is how it looks on Tiktok:

Here is how it looks on Google:

Mistake 4: Never Optimize Based on Daily Results

One critical mistake that many advertisers make, especially advertisers that I meet personally, is the horrific mistake of optimizing based on daily results. This is wrong for many reasons, but the primary reason is that daily results mean nothing. Absolutely nothing. Daily results are just a grouping of 24 hours, and making decisions based on 24-hour performance can lead to suboptimal strategies. For many reasons, daily performance can fluctuate significantly due to a variety of factors, including user behavior patterns, external events, and random chance. Optimizing based on such a short timeframe ignores the broader trends and can cause advertisers to make hasty decisions that do more harm than good.

Purchasing behaviors vary widely, with users often converting days after their initial interaction with an ad. Evaluating performance based on daily metrics ignores these delayed conversions, leading to misguided optimization decisions. Instead, it’s crucial to consider longer periods when analyzing ad performance to get a more accurate picture of how well your campaigns are performing.

Advertising algorithms also do not optimize based on daily results. They use lifetime performance data to make adjustments. When you set a daily budget, you're actually indicating how much you want to spend per month. The algorithm then distributes this budget over the month, not focusing on daily fluctuations. This means that looking at daily results can be misleading and cause unnecessary panic or over-optimism based on short-term data.

Instead of optimizing daily, consider longer evaluation periods. Some advertisers may choose 5-7 days, while others with larger budgets might opt for weeks or even months. The key is to avoid the pitfall of daily optimization and focus on broader performance trends.

Daily optimization can lead to overreaction to short-term fluctuations. For example, if you see a dip in performance on a particular day, you might be tempted to make immediate changes to your campaign. However, this reactive approach can disrupt the overall strategy and lead to suboptimal decisions.

By focusing on longer evaluation periods, you can identify more reliable performance trends and make data-driven decisions. This approach allows you to see the bigger picture and understand how your campaigns are performing over time.

Additionally, longer evaluation periods help smooth out seasonal variations and other external factors that might impact daily performance. For example, weekends, holidays, and special events can cause fluctuations in user behavior. By considering longer periods, you can account for these variations and make more accurate assessments of your campaign's effectiveness.

To implement this approach, establish a regular schedule for reviewing and optimizing your campaigns. Instead of making daily adjustments, review performance data weekly or bi-weekly. During these reviews, analyze key metrics such as conversion rates, cost per acquisition, and return on ad spend. Look for patterns and trends that provide insights into your campaign's performance.

Conclusion

Successful advertising requires more than just choosing the right platform. It involves a strategic approach that considers long-term performance, accurate attribution, and meaningful optimizations. By steering clear of these common pitfalls—running ads with view-through conversions, enabling partner networks as a placement, using different attribution windows, and optimizing based on daily results—you can significantly enhance the performance and profitability of your ads on Meta, Tiktok, and Google.

By focusing on these key areas and making informed decisions, you can ensure that your ad budget is used effectively, driving real, measurable results for your business. Remember, successful advertising is a marathon, not a sprint. Take the time to analyze, optimize, and refine your strategies for the best possible outcomes.