John Lee, Author at Flightpath

How to successfully advertise without third-party data

April 29, 2022

By John Lee

An artistic representation of using first party data for audiences.

After all the talk clears and the dust settles on Apple and Google’s third-party data changes, what will become of digital marketing? Unlike some have led us to believe, this won’t be the end of digital marketing. It won’t even be the end of digital marketing as we know it. It does, however, mark the beginning of new and innovative ways to find customers who are ready to listen to, and act on, the messages we share.

Here are a few points to show how the shift in digital marketing will affect some initiatives, and make room for new ideas.

Thinking deeper about first-party data

Using first-party data is by no means a new tactic, but without the amount of third-party data we all once had access to, this first-party data should be looked at in new and more clever ways.

This may be as simple or complicated as optimizing your website to glean more information from site visitors. Do you have tracking in place to know who’s interested in your products, enabling you to better message and respond to them on other digital platforms?

These visitors are already part-way down the sales funnel, and may only need a small bump to take them all the way. While it may seem like work up front to optimize your website for a more robust tracking system, the payoff for finding stronger-interested parties can be worth it.

Creating smarter lookalike audiences

The science of data is here to stay, and the implications for what it can do are immeasurable. Algorithms are being created every day to pull in data sets and produce predictive responses to numerous questions around the world—and advertising is no different.

Like most algorithms, the more data points that can be factored in, the better the outcome. Or in advertising terms: the better the lookalike audiences that are created. In addition to strong lookalike audiences, it’s important to find the right platform for those audiences to be targeted.

For example, a lookalike audience for a healthcare company marketing to insurance providers may be useful on LinkedIn, but not as useful on Reddit. Whereas a specialty sneaker company might easily find a strong audience on Reddit, but not on LinkedIn.

Allowing ideas that might fail

Before you completely fight this idea, hear it out. In no way are we proponents of bad ideas just for the sake of trying anything. But we are proponents in making calculated risks that might just win big—and allowing trials before jumping full in.

One of the greatest things about so many forms of digital marketing is that they are easy to try, track and improve. There are countless times where simple things like A/B testing can prove helpful to finding the strongest social media ad, the better ROI for SEM, or the landing page design that produces the most leads.

With third-party, first-party or lookalike audiences, it can sometimes be the unexpected message that resonates best. Or the outside-the-box offer that adds brand loyalty from current customers—further strengthening the first-party data already accessible to a company for marketing.

When is Google Removing Third-Party Cookies?

August 23, 2021

By John Lee

It’s been months since Google first announced they’d phase out third-party cookie tracking from their platform—or as one news outlet named it “the cookiepocalypse.” But since their internet-breaking announcement, we’ve yet to see it actually happen. Just what is the holdup on this plan, and what does it mean for your company, both now in 2021, and for later years?

What are Third-Party Cookies?

First, let’s set the baseline on just what third-party cookies currently do. You may have noticed the cookie disclaimer that pops up when visiting websites for the first time. In the simplest terms possible, it helps the site keep track of what you’ve been doing on the site.

Agreeing to website cookies can be a useful thing for user experience. It’s how a website remembers your language preference, what country you’re visiting from, or how a shopping cart remembers what you put in it the day before. It also helps a company see things like user flows and popular pages.

Third-party cookies mean someone outside of the company that owns the website might be tracking your data too. They don’t typically track things quite as refined, but they may have something on the homepage just to let them know you visited.

The biggest way they use that data is to retarget ads based on your preferences. One of the easiest examples of this is when you visit a shoe site, and almost immediately after you see ads on Facebook for that same shoe site, or others they think you may like as well.

Why is Google Removing Third-Party Cookies?

Some browsers, like Safari and Firefox, already limit third-party cookies. The reason Google’s announcement is so newsworthy is that they have the lion’s share of users, so their decision to cut data has larger repercussions on the advertising industry.

But so far, no real updates have been made. The initial news said mid 2021, then there was news that they were planning to slowly fade out the options. And now it seems the plan may not be in effect until 2022.

So, how much push is coming from Google to remove third-party cookies, and how much is coming from industry pressure regarding privacy policies? The answer can depend on who you talk to.

How Can We Track Usage Without Third-Party Cookies?

Although the changes haven’t happened yet, agencies around the world are already prepping for what things might look like without the ability to advertise as we’ve been doing for years.

Google, for one, has already stated that they are coming up with their own ways to track usage. But what they’ll do exactly remains to be seen. Will their updates push more automation from their own platform? If they keep more data within their own site, it will force companies to use only their platforms, and may result in less ability to pinpoint more nuanced audiences.

Another plan is to rely more on what Google has coined FLoC (Federated Learning of Cohorts). Proponents of this method say it’s a more privacy-first way to advertise, but opponents say we’ll lose a lot of structure that’s needed to keep ROIs where they are now.

So what is FLoC? It’s a way to target based more on basic profiling; relying mostly on audience groups. It would be a more contextual way of advertising. More like how traditional ads of the past worked, like TV and magazines. You may not know your audience went to the exact shoe site, but based on demographics and other information you can make an educated assumption they’ll be interested in the shoes you want to advertise.

Because of the way this will most likely work, retargeting will be limited, and will be limited to Google’s cookies for information and advertising.

How much will this affect ROI? How well will FLoC work to target smaller audiences? Will creatives need to change to adapt? All of these questions are still in constant debate, and we won’t know answers for sure until we see the new plans in action.

What we do know is that similar to all digital advertising, this can be approached with a strong strategy and a readiness to quickly adapt and change as needed to reach appropriate audiences and guarantee positive outcomes.

Are you Getting the Most Out of Search Advertising?

May 13, 2021

By John Lee

One thing is certain from this last year: online spending is not going anywhere. Which is why through all the ups and downs other industries faced in 2020, search advertising has raised expectations for spending this year.

Spending for all touch points, which combine desktop, mobile and tablets, is forecast to increase by at least 10%, and as much as 25% in some areas. That makes it the fastest growth since 2018. This is good news for anyone looking to increase conversions for your brand.

How did the pandemic affect search advertising?

When news of shutdowns first hit, uncertainty was high, and advertising responded in kind. In Q1 and Q2 many advertisers took a cautionary step and lowered ad spending. But as more people began to look around online, and order things directly to their doorstep, things changed. While advertising spending overall will likely increase from some of the flattened numbers of 2020, search advertising is forecast to do especially well.

How can search advertising increase conversions?

Like any good marketing, the best way to reach your audience is where they’re already at. And right now everyone is online. Increasing conversions through search engine marketing works with a consistent strategy and well-researched keyword usage.

Targeting audiences through the right keywords can be taken a step further by combining other targeted points like website visits and interests or behaviors. This use of custom audiences can make sure your ad spend is working its hardest to reach the correct users.

Various conversions can exist within the user journey, and each should have a desired CPA. Campaigns or ad groups defined by those different users should be segmented as such and can be organized by different keyword types: priority keywords, head terms (low hanging fruit), longtail, impression share, and conquesting (competition) to name a few.

How can search advertising increase sales?

Targeting your audiences for clicks or conversions can include a variety of strategies. The top way to optimize is by dividing into useful campaign/ad group segments, which are based on keyword groupings. With these divisions made, the next step is to use separate bid strategies for each grouping. These would ideally have Target CPA’s set to control costs.

These groupings allow keyword research, audience segmentation and strategy to work together. When everything is properly in place, your advertising will find ready audiences with the type of communication that will get their attention the best.

Updating Facebook Ad Performance for Apple’s New iOS

January 20, 2021

By John Lee

Apple’s upcoming advertising changes are rolling out soon, and along with that is Apple’s Private Click Measurement (PCM), giving users more options to opt out of the data-tracking methods that Facebook has been using to target audiences.

While some parts of this rollout are seen as a step in the right direction for privacy, it can mean different things for advertisers like us, and the companies that are aiming to reach the right audience with their ads. Because of these shifts, it’s important to strategize updates and changes that ensure ad dollars are still being used wisely.

Upcoming Roadblocks

Two of the biggest problems Facebook has pointed out are app-to-web conversion measurements and cross-domain measurements. For app-to-web, the problem we’ll start to see with conversions is that tracking within the app can no longer be recorded. For example, if your user clicks on a link within an app, like Instagram, it typically opens a browser window still within the app. This type of user experience can no longer be recorded on an iPhone because it is within another app.

The other experience that will no longer be tracked is for companies that have a URL redirect. If a user clicks on a link, but that link redirects to a different webpage, that will be the end of tracking.

While this may seem like only two small situations, ultimately it can become a huge issue if not addressed properly. Performance will be harder to achieve due to lack of data, current Facebook algorithms may not be as strong, leading precise targeting to become a concern.

Solutions for a Smooth Transition

Because Facebook feels these types of experiences are important to ensure users are given ads they care about, they are building their own Aggregated Event Measurement (AEM) to work within Apple’s new protocols.

Although this will allow advertisers to access user data for better targeted ads, there are limitations and updates that need to be made. Members of Flightpath’s marketing team recently participated in a Facebook webinar that outlined these new protocols:

Eight-event limitations for pixel setup
Conversion window updates
Reduction in custom audience sizes
Value optimization within Events Manager

While Facebook did not give point-by-point instructions for how to adapt to all of these changes, we are developing strategies to compensate. One thing we will be doing is making required updates in a timely manner to mitigate any foreseeable bumps in user data. As with all advertising, it’s very imperative to us that these changes are watched closely as they roll out, and that we are making swift adjustments as necessary to ensure the best ROI for all ad dollars spent.

For anyone working within Facebook’s platform, here is a quick to-do list to make sure you are on top of any pending changes with the new Apple iOS program.

Verify your domain in Facebook Business Manager
Prepare to only have 8 events per domain, prioritizing events as needed
Anticipate changes to attribution windows, and update automated rules if necessary
Identify campaign optimization strategies that may require testing (alternative audience options or bidding strategies) as needed

As we said at the start, many of these changes aren’t ideal for the ad targeting we have been able to do to this point. But with proper strategies in place, and the ability to adapt to new updates as needed, we can all continue with useful ads targeted to the right audiences.

Facebook Brand Lift Overview

October 22, 2020

By John Lee

If you advertise on Facebook, or plan to in the future, you may have considered doing a Brand Lift test to see deeper results of how that advertising is supporting your initiatives. But it’s important to note that not all brands need or have access to Brand Lift studies. Our experience working with brands who have done them has given us useful insight into how they work, and who they work best for.

Brand Lift Basics

The point of a Facebook Brand Lift test is to use polling, aimed randomly at Facebook users who have seen your ads, to learn more about ad recall and effectiveness. It can be done for one specific campaign, or across a breadth of advertising within specific dates. The test will run within the dates of the campaign, or after it has completed, but not both.

After determining the audience group, Facebook will randomly create test and control groups, polling them with quick surveys asking questions ad and brand recall.

Who They’re Useful For

Brand Lifts are useful for companies planning to run one campaign for a single brand or product. If you have multiple offerings or products within your company, and advertising is mixed or crosses products, this type of study may not produce clear enough data to make it useful.

Another thing that is required for these tests to work is that creative and budget remain the same throughout the campaign. Again, switching things or testing messages on a campaign can be valuable for other results, but would not yield the data that a Brand Lift is meant for.

Facebook has also set company requirements for a Brand Lift test. One is that in order to run this free test you need to already be working with an Account Representative, and possibly meet certain budget requirements for the advertising you are testing. If you’re not sure if you qualify for a Brand Lift test, we can let you know!

Outcomes to Expect

A well-planned Brand Lift test will be able to help you understand if your advertising is memorable, how aware it has made people of your brand or product, and if potential customers are considering purchasing what you offer.

The results can be very effective in planning future ads and messaging strategies to best reach future customers, grow an audience or increase sales.