Blog

How to Use Customer Segmentation to Improve the Performance of Your Marketing Campaigns

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on How to Use Customer Segmentation to Improve the Performance of Your Marketing Campaigns

How to Use Customer Segmentation to Improve the Performance of Your Marketing Campaigns

Your audience wants personalized marketing from your business.

In fact, they expect it. According to research, 71 percent of customers expect businesses to send them personalized marketing messages, and 76 percent are disappointed when they receive generic communications instead.

The challenge? If you don’t know your audience, you can’t send them personalized content. You don’t know what matters to them, so you can’t reach them on the right level.

If this dilemma sounds familiar, don’t worry. I have a solution for you, and it’s called customer segmentation. Customer segmentation helps you understand your audience so you can target your marketing campaigns with greater precision. Let me show you how it works.

What Is Customer Segmentation?

Customer segmentation means dividing customers into groups, or “segments,” based on traits they have in common such as age, buying habits, gender, and needs.

Businesses use customer segmentation models to better understand their prospects so they can target them with relevant personalized marketing campaigns including ads, emails, and social media posts.

Customer segmentation isn’t just about reaching a new audience more effectively, though. It’s also a way to reconnect with lapsed customers and encourage new purchases by sending them carefully targeted messages.

Remember, every customer is unique. They each have own buying behaviors and reasons for choosing you over your competitors. While it’s impossible to personalize your marketing to every individual, a customer segmentation strategy is the next best thing.

Why Is Customer Segmentation Important?

For one thing, it helps you improve your customer service. By understanding your customers’ needs and wants, you’re better placed to help solve their problems.

Does customer service matter? Absolutely. Research says one in five customers will abandon a brand after just one poor customer experience, so the more effort you invest in great service, the better.

Similarly, segmenting your audience helps build customer loyalty. How? Because customers are typically more loyal to brands offering personalized messaging—for 79 percent of consumers, the more personalization a company uses, the more loyal they are.

What do loyal and happy customers have in common? They’re more likely to shop with you. By personalizing the shopping experience through segmentation, you create more dedicated customers, so you increase conversions over time.

Not convinced? Well, studies show that over 60 percent of customers are likely to be repeat buyers after a personalized shopping experience, so the stats speak for themselves.

Customer Segmentation Models

You can use various customer segmentation models, depending on your business needs and marketing goals. Here’s a look at seven of the most common models.

1. Demographic Segmentation Model

Demographic segmentation means dividing people into groups based on certain demographic factors, including age, income, marital status, and occupation.

Let’s say your audience is men and women aged between 30 and 65. You want to run a TikTok campaign to promote a new product.

61 percent of TikTok users are women. 11 percent of users are over 50.

If you only run a campaign on TikTok, you miss out on a huge chunk of your target audience. Perform some demographic segmentation, and you’ll know to target Facebook, too, since 73 percent of 50- to 64-year-olds use this platform.

Want to try it?

Set your campaign goal. Choose your variables, whether it’s age, gender, and so on. Select your platforms to run personalized marketing campaigns, such as social media, email, etc.Measure success using tools like Google Analytics and revise your campaigns as needed.

Pros and Cons of Demographic Segmentation

On the plus side, it’s easy to use this model, and it helps you adjust your tone to target different genders and ages.

The main downsides? You risk making false assumptions about a particular segment. You could also lose your brand voice by targeting such varied demographics.

Always use this customer segmentation model alongside other techniques. For example, it might be helpful to know a customer’s buying habits and values, or where they live.

2. Geographic Segmentation Model

With geographic segmentation, you categorize your audience based on where they work, live, and shop.

This type of customer segmentation analysis is fairly straightforward. The main disadvantage? Ironically, it’s simplicity. On its own, geographic segmentation doesn’t reveal much about your audience, but you can use it alongside other models on this list to build the fullest possible picture of your audience.

How to Segment Customers Through Geographic Segmentation

Here’s how to get started with geographical segmentation:

Determine your segments. You can divide people by, for example, climate, culture, language, or land area. Gather data, such as website location data and sales data, to identify the size of your community. Send targeted messages to customers based on these segments. As an example, you might run paid ad campaigns based on location, or if you’re launching an exclusive location-based product, email your target audience a promo code.

Case Study: McDonald’s

McDonald’s frequently uses geographic segmentation to target different audiences around the world. For example, here’s a burger found in McDonald’s India:

McDonald’s creates products to suit its diverse audience and tap into the flavors and products they may respond to based on geography.

This brings me to another advantage of geographic segmentation: exclusivity. Since the McDonald’s menu varies by location, each item feels exclusive, harder to acquire, and more valuable, which may increase conversions.

3. Psychographic Segmentation

We each have unique personalities, but we share traits or characteristics. Psychographic segmentation means forming groups based on common traits such as hobbies, lifestyle choices, personality traits, cultural beliefs, and values.

Psychographic segmentation helps you understand a customer’s psyche so you can devise highly focused, relevant campaigns. However, the main challenge is gathering (and organizing) the relevant data.

How to Use Psychographic Segmentation

Follow these steps to start using psychographic segmentation:

Determine your ideal customer. Who are you selling to? What do they love about your products? This stage may involve some consumer research. Choose your segments, such as hobbies, values, or personality traits. Identify where your audience congregates. For example, over 1.5 billion people visit Reddit every month, and 38 percent of Americans listen to podcasts every month. Perform some (more) consumer research. Whether you run Instagram polls or send surveys, ask your audience what type of content they want from you. Evaluate the data to decide how to properly target your groups.

Case Study: Patagonia

Patagonia, an outdoor clothing brand, knows its customers care about sustainable living. They’ve made sustainability a core part of their brand messaging:

If you ran a store like Patagonia, you could segment customers based on whether they prefer hiking or cycling and then send targeted campaigns to meet their needs while retaining this core brand message.

4. Technographic Segmentation

Technographic segmentation means categorizing people depending on the devices, hardware, and software they use. Why does this data matter? Well, according to statistics:

79 percent of U.S. smartphone users purchased something online through their mobile phone in the last six months.40 percent of consumers switch to a competitor after one (yes, one) bad mobile phone experience. Purchases made on tablets are set to rise to over $64 billion in 2022.

As a marketer, you should care about how people are accessing your content so you can optimize their user experience (UX) and target them effectively. Technographic segmentation can help.

How to Perform Technographic Segmentation

There are a few ways to segment your audience using this method, but here’s how I suggest you start.

Know your audience: Identify your customers, as they will determine which categories you choose. Pick your segments: For technographic segmentation, you might group people based on the devices they use, the software they’re working with, the apps they prefer, or how they use technology. Gather data: Collect the data you need to segment customers. You might do this by scraping websites, sending surveys, or even purchasing data from service providers.

Armed with this data, you can create your campaigns.

Example of a Technographic Segmentation Campaign

Let’s say you run a tech store. Some customers use Norton 360 for PCs. Others use Avast Security for Mac.

You split your marketing campaign by software. You send one email to Norton subscribers offering a discount on their annual subscription. You send another email to Avast customers offering the same discount for Avast.

The result? Emails that speak to your audience’s specific tech needs, which increase your chance of making conversions.

You could take it further, too. Say, through analytics, you notice your Norton PC customers are looking at mobile antivirus solutions. You could send them a discount code like this one from PCWorld:

By anticipating what matters to your audience based on their tech preferences, you’re meeting their needs…and hopefully nurturing them through to checkout.

Is this a perfect customer segmentation model? No. One significant drawback is its limitations: Knowing a customer’s tech preferences is only one part of what shapes their buyer’s journey. However, it’s a marketing technique worth adding to your toolbox.

5. Behavioral Segmentation

Want to know how your audience interacts with your business? Try behavioral segmentation.

Behavioral segmentation means grouping people together based on behavior patterns. These patterns reveal how consumers feel about your business so you can determine how to successfully reach them at every stage of the buyer’s journey.

As with other models, behavioral segmentation can be used at any point in your marketing strategy, whether it’s to revamp a landing page or send promotional emails.

How to Use Behavioral Segmentation

First, identify the behavior patterns to track. There are many ways to approach this, but you might segment customers based on their:

buying stageengagement historical purchase historypurchase frequencyresponse to previous marketing campaigns

For example, say you group customers based on engagement. What counts as an “active” and “lapsed” customer varies depending on your business, but here are three groups you might have:

Active customers shop with you every month. Infrequent customers only buy products every few months.Lapsed customers haven’t purchased from you in a year.

Next, you can devise three separate marketing campaigns. You might send active customers a loyalty discount, and infrequent customers a separate discount to tempt them back.

Once your campaigns are up and running, track your analytics. If you’re not getting the results you want, adjust your campaigns and try again.

Netflix and Behavioral Segmentation

With over 221 million subscribers, Netflix knows how to use behavioral segmentation to satisfy customer demand.

Netflix uses machine learning to track what customers watch. The algorithms generated help Netflix customize everything for each customer, from the homepage to the show recommendations.Netflix can use A/B testing to track the impact of different recommendations and personalization features.

Behavioral segmentation has a significant downside, though: There’s always the chance you get the algorithms wrong. That said, if you track results diligently and respond to your findings, you can offset this drawback.

6. Needs-Based Segmentation

Successful marketing often comes down to showing prospects how your goods or services meet their needs. That’s where needs-based segmentation comes in.

With needs-based segmentation, you’re grouping people based on what they need from your product. The benefits they’re looking for when they buy something. What pain points they have, and the problems they need solving.

The biggest challenge? Identifying what these needs are.

For example, say you’re a food brand. Two prospects follow you on social media. One cares about fresh chicken, and the other wants vegan food. You might sell meat and non-meat products, but the same ad campaign won’t appeal to both.

Driving down into groups’ needs and motivations helps you maximize your campaigns.

Let’s do a simple comparison. Heck sells gluten-free vegan and non-vegan meat. They know some customers love the gym and care about high-protein snacks, so they launched a campaign to sell their meat at local gyms:

They know other customers care less about fitness and more about a vegan lifestyle, so they frequently create social media posts around meat-free products:

Heck clearly spent time learning about its wider customer base and what drives them so it can effectively reach every segment while retaining a consistent brand voice.

Here’s another example. Beauty store Revolution lets customers shop by skin concern and by ingredient to directly target consumers’ needs:

Needs-Based Segmentation Pros and Cons

Now that you understand how this customer segmentation model works, is it right for you?

Well, there are clear advantages. Needs-based segmentation helps you market with greater accuracy than, say, targeting groups by age or location. It’s comprehensive and effective, and it could help you build loyal customer relationships.

The main drawbacks? It’s challenging to identify the “right” needs to target, and if you don’t have accurate data, your campaigns may fail. What’s more, consumer needs evolve, so you’ll need to review your strategy regularly to maximize your campaign effectiveness.

How to Perform Needs-Based Segmentation

Here’s the simplest approach.

Start with your products or services. Look at them from every angle and write down all their features and benefits. Build customer personas around these features. If you know how to segment customers based on behavior, age, location, etc., use the data you already have to help here. Finally, reach out to customers and learn what matters to them. You might, for example, look at product reviews, ask for customer testimonials, or send out questionnaires.

Once you have enough data, use your findings to create segmented marketing campaigns. Track your campaigns and tweak them as needed.

7. Value-Based Segmentation

The better you understand how much it costs to lose a certain client’s business, the better you can direct your marketing efforts. Value-based segmentation can help you by grouping customers together based on their value to your business.

Why group customers together this way? Well, there are two advantages.

Firstly, if you know which customers spend the most money on your products, then you know which customers you can’t afford to lose. You can direct resources into providing these customers with highly targeted campaigns and great customer service.

Secondly, you can identify your most loyal clients and how much it costs to retain their business. Once you know a customer’s relative value, you can decide if it’s worth retargeting these inactive customers with personalized messaging.

Is retention worth the effort, though? There’s evidence that it can be up to seven times more expensive to acquire rather than retain customers, so yes, retention matters.

Using Value-Based Segmentation

Here’s how to segment your customers on a value basis.

Decide on your campaign goals. Maybe you want to identify your most lucrative audience and launch an ad campaign for your high-end products, or you want to nurture lapsed customers back to your store with enticing loyalty discounts. Identify your segmentation criteria. For value-based marketing, you might segment customers based on average spend or relationship duration as described above. Determine how you’ll target customers based on your findings; for example, on social media, by email, or through paid ads. Analyze your efforts such as by running regular A/B testing or asking customers for feedback.

On the plus side, value-based segmentation helps you quickly identify your most valuable customers in order to target them more effectively. However, if you’re a startup or young business, you may not have enough relevant data to use this customer model just yet.

Case Study: Global Cruise Company

Here’s an example of the basic value-based segmentation principles in action and how this method helps with retargeting and conversion.

Merkle, a marketing company, helped a global cruise company develop a value-based approach to their next marketing campaign.

The cruise company sent the same messages to every customer regardless of their lifetime value (LTV). To boost revenue, they wanted to segment customers based on their LTV to send tailored ads and emails.

The company broke down each customer’s total predicted economic value. Once they identified the highest-value and most loyal customers, they could better nurture them through the sales funnel with specific, smaller campaigns.

The results? Five percent of lapsed but loyal customers returned, and they shortened the purchase cycle by 24 percent. All it took was some focused, personalized messaging based on a customer’s relative value.

Customer Segmentation Frequently Asked Questions

What tools do I need to do customer segmentation?

You need data to segment customers effectively, so you’ll want analytics tools such as Google Analytics. You might also use dedicated customer segmentation software, depending on your budget and business goals.

Is customer segmentation worth it?

By segmenting your customers, you learn more about your target audience and what matters to them. The result is more effective marketing campaigns based on the unique needs of each segment within your broader audience base.

What type of campaigns does marketing segmentation work best with?

Segmentation works best on any channel when you’re using personalized ads aimed at certain people because you can run multiple smaller, highly targeted ad campaigns designed to deliver the right message to the right audiences.

How is customer segmentation used in customer retention?

Customer segmentation ensures your existing customers don’t feel overlooked. You can segment your loyal customers into smaller groups to deliver relevant, loyalty-based rewards that could help increase customer retention over time.

{
“@context”: “https://schema.org”,
“@type”: “FAQPage”,
“mainEntity”: [
{
“@type”: “Question”,
“name”: “What tools do I need to do customer segmentation?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: ”

You need data to segment customers effectively, so you’ll want analytics tools such as Google Analytics. You might also use dedicated customer segmentation software, depending on your budget and business goals.


}
}
, {
“@type”: “Question”,
“name”: “Is customer segmentation worth it?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: ”

By segmenting your customers, you learn more about your target audience and what matters to them. The result is more effective marketing campaigns based on the unique needs of each segment within your broader audience base.


}
}
, {
“@type”: “Question”,
“name”: “What type of campaigns does marketing segmentation work best with?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: ”

Segmentation works best on any channel when you’re using personalized ads aimed at certain people because you can run multiple smaller, highly targeted ad campaigns designed to deliver the right message to the right audiences.


}
}
, {
“@type”: “Question”,
“name”: “How is customer segmentation used in customer retention?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: ”

Customer segmentation ensures your existing customers don’t feel overlooked. You can segment your loyal customers into smaller groups to deliver relevant, loyalty-based rewards that could help increase customer retention over time.


}
}
] }

Conclusion: Customer Segmentation

If you’re trying to upgrade your marketing, customer segmentation is your friend. By segmenting your audience, you can learn what matters to your customers, run targeted, more effective campaigns, and ultimately convert more leads into customers over time.

Start by evaluating the customer segmentation models I’ve described and consider which combination works best for your business goals. If you need any guidance for choosing between customer segmentation types, though, check out my consulting services to discover how my team can help.

Have you created your customer segmentation strategy yet? Which model do you find works best?

What is your least favorite part of PPC? [POLL]

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on What is your least favorite part of PPC? [POLL]

Love it or hate it, PPC isn’t going anywhere. At least not for a while.

As PPC experts, we get to deal with everything from new platforms, added features, disappearing metrics (we’re looking at you, Facebook), angry clients, privacy issues and baffled relatives who ask “what do you do, again?”

For example, doesn’t it just frustrate you to no end that Facebook got rid of nearly all of their targeting options? The exact thing that made Facebook a goldmine for businesses in the past is the exact thing that’s forcing advertisers away from the platform in 2022. Ugh!

But we want to hear it directly from you, whether you do PPC on the frontline at an agency, in-house, or as a freelancer.This is your chance to let us know: What is your least favorite thing about PPC?

Be honest: tell us what you find frustrating, difficult, tedious or even downright painful – and why. We may include your answer in our write-up of the poll results.

Submit your response below. We may include your quotes in a write-up of the poll results!

The post What is your least favorite part of PPC? [POLL] appeared first on Search Engine Land.

WebPageTest’s new Opportunities and Experiments: test practically anything

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on WebPageTest’s new Opportunities and Experiments: test practically anything

WebPageTest’s new Opportunities and Experiments: test practically anything

There’s never been a better time than now for a web developer’s approach to SEO.

The pace at which tools and resources, both new tools and anything familiar, innovate and open options for us also demand that we keep up. Recently, that has meant more requirements for performance optimization as Google releases algorithm updates and changes to metrics calculations.

One tool that you should be familiar with is WebPageTest. They recently released some incredibly useful new fully integrated test tools.

WebPageTest now proxies real-time user-specified HTML modifications through Netlify to run comparison tests right inside their user interface. No coding is required.

Genius makes sense

Smart application logic in three huge areas of concern bubble findings up for you, but not just with text blurbs, with re-test options prepped for you to run combinations of variations for comparison. The array of tests available in WebPageTest now means there will be no more setting up tests using third-party proxy tools that duplicate what you can test directly.

This was all technically possible before and the original approach continues to have importance.

Although impressively comprehensive, there will always be tests you will need to run using a proxy host of your choice. This requires handling JavaScript and Cloudflare, however.

With WebPageTest you get to point and click.

Pesky lab data

Always keep in mind the best possible combination of numbers from lab tests may not yield the same numbers in the field. It can actually result in broken website features.

Scripts and styles have developer-defined load order where any change can mean a breaking change that is not suitable for production. A proxy host can provide access for QA as part of the optimization process.

With that warning out of the way, let us tell you how great it is having a testbed for demonstrating HTML optimizations. It’s been the basis of our workshops and conference sessions now for well over a year.

Our Search Engine Land guide articles can help you set up a testbed. We’ll be using an updated version at SMX Advanced. Join us live if you can make it.

Opportunities

WebPageText’s Opportunities text is available to everyone in reports.

You won’t need JavaScript skills to run HTML variation comparisons anymore. Instead, you will need a paid account to run built-in proxy tests labeled Experiments.

The free account gives you better access to reports and history, but not running Experiments. You can still write JavaScript and proxy your own tests for free.

It’s just nowhere near as handy and it takes up way too much time.

Free opportunities. Paid account for experiments.

Experiments

Select the Opportunities & Experiments menu item in a WebPageTest report and you will be presented with a comprehensive list of findings.

Opportunities here are derived from real-world test conditions (simulated with hardware where possible). Our test indicated an opportunity to re-test experimenting with render-blocking resource variations (typically JavaScript and CSS), lazy loaded images, self-host third-party script and much more.

Pro Account Required for Variations

Test async, defer, or even inline scripts and stylesheets using the interface. We’ve been writing Cloudflare Worker JavaScript to proxy these tests and we also added inline style rules to defer loading content towards the bottom of the page, including the footer. The initial array of WebPageTest integrations can handle most, but not all, of our original tests.

It’s a snap to put tests together now.

Modify test settings and start running variations to hone in on the holy grail of green Core Web Vitals across the board. The offering is amazingly comprehensive and covers far more than what affects a webpages’s performance.

You’ll find three categories which group opportunities to experiment by the following questions:

Is it quick? Quickness categorizes and groups performance optimization experiments.Is it usable? Usability groups HTML validation errors that can mess with screen readers and things that affect layout shifts.Is it resilient? Resiliency goes to security concerns including mixed protocols. Modify test settings with a checkbox interface and start running variations. You will get refined options in the comparison report.

Dashboard for a test suite

WebPageTest has to resemble a dashboard for a test suite and manage to do that inside reporting that provides more detail than Lighthouse, and with far better waterfall chart representation than Chrome Dev Tools.

Although it’s true with point and click you can run HTML experiments in a “no-code” environment, the detail provided and navigation requires experience – and coding experience is best.

A new built-in Experiment replicates another Cloudflare worker task by removing all JavaScript. Having stuff like that so accessible is exceptionally handier than writing a script for test variations.

Advanced Experiments allow us to insert HTML in key locations, test tactics to change load order, fail to load, or modify, including minify, resources.

There’s literally nothing technically stopping us from testing practically anything on any page.

Fall into the pit of success

Comparison reports themselves serve to funnel you into selecting and re-testing more variations. The result metrics banner includes color-coded improving and worsening scores between control and experiment.

A remaining opportunities section with a subset of experiment switches appears below. You can click your way through to significant improvement.

We’ve done the hard work to write tests for demonstration at SMX Advanced and when we’re live you can expect us to cover this major update to the very tools we used. It’s going to be so much easier.

We will see if the rapid text cycle of WebPageTest experiment integrations gain what we were already preparing to deliver. Let’s see if we can get to green across the board.

The post WebPageTest’s new Opportunities and Experiments: test practically anything appeared first on Search Engine Land.

Meta rumored to develop ‘Basic Ads’ in response to recent privacy changes

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on Meta rumored to develop ‘Basic Ads’ in response to recent privacy changes

Meta Platforms’ Facebook is reportedly in the early stages of planning a Basic Ads product. Basic Ads would be aimed at advertisers who are looking to build awareness around their brands. 

What are Basic Ads?

A Basic Ads product would offer and report only the simplest metrics such as engagement and video views.

With the release of iOS14 in 2020, and the option for users to opt-out to having their data collected, Facebook advertisers are having a more difficult time reporting on performance.

The initial response

Since Meta hasn’t officially announced the new product, little has been said about it in the advertising community. But Facebook veteran Curt Maly of Black Box Social Media, who is aware of Meta’s plans, said it’s interesting to think how a basic product with no targeting and no objectives would be beneficial. 

“More than 90% of all the online marketers I know are focused on direct response where efforts can be directly tracked… with this basic ads platform, it appears that tracking will be rather difficult,” Maly said. “Branding and awareness are used by larger companies with deeper pockets, most small businesses can’t compete with a brand who spends money ‘branding,’ small business owners need/want to drive results now.”

He added that brand new online advertisers and ad vets flock to Facebook to see fast results from targeting, conversion objectives and tracking results.

“If these three major goal lines are moved, I think we will be seeing a lot more people flock to Google/YouTube ads, TikTok ads and Apple’s new ad platform,” Maly said. “I mean Apple didn’t update iOS to ‘help protect users.’ Apple collects all the info they block on Facebook, Apple is about to get into the ad platform game once again and this is yet another reason for people to flock to a better ad platform.”

Privacy changes impact on Meta revenue

Meta estimates a $12.8 billion hit to their revenue in 2022 due to the Apple changes. Meta CEO Mark Zuckerberg also reports “unprecedented levels of competition” from new platforms such as TikTok.

In April, Meta laid out a three-tier plan to save its ad business in the wake of Apple’s privacy crackdown. At the time, Meta COO Sheryl Sandberg described the strategy as “doing more with less data.”

Meta’s response

Facebook declined to comment on the new product. There has not been a release date for a rollout, though Business Insider reports testing to start in the EU ahead of the U.S.

What this could mean for advertisers

The Facebook ads platform was previously known for its wide range of targeted audiences and demographic options. As of late, it seems like a lot of advertisers and businesses are moving away from the platform, reporting a decline in performance.

Basic Ads could be a good alternative if they’re able to follow GDPR regulations while still providing useful targeting options.

Basic Ads may work well for household names such as Nike or Netflix whose goals are engagement and awareness. But smaller businesses that rely on more granular targeting and lead generation, such as courses for business owners, or youth soccer camp registrations, may have a more difficult time and may abandon Facebook for good. 

The post Meta rumored to develop ‘Basic Ads’ in response to recent privacy changes appeared first on Search Engine Land.

How your PPC conversions will be impacted without privacy-first measurement

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on How your PPC conversions will be impacted without privacy-first measurement

Within the next 12-15 months, third-party cookies will retire across digital marketing channels.

Savvy advertisers know they need to begin developing a game plan for the cookieless future, but what will happen to those who don’t adapt to these changes?

Above all, marketers will suffer from signal loss, which will negatively impact how we measure campaign performance, optimize campaigns over time, create audiences for ad distribution and drive growth within our digital channels. 

The industry sea change with the lion’s share of attention is the retirement of third-party cookies in Google Chrome.

Sure, other browsers, including Microsoft Edge, Apple Safari and Mozilla Firefox, have previously restricted third-party cookies. Chrome is more monumental simply because of its market share.

SimilarWeb recently released a study that showed Chrome was the world’s most popular browser with 62% of web traffic. 

To recap from my previous article, Google Chrome will retire third-party tracking cookies around Q3 2023. That is an approximate timeframe for this monumental change, but it gives us a target to make sure that our digital marketing campaigns will be ready.

This might sound like the distant future, but many of the measurement solutions needed to replace the functionality of third-party cookies could require significant time and effort from development teams.

This type of support usually requires a few cycles to be prioritized on project roadmaps.

Getting started in the next couple of months will be beneficial in the long run.

Look at it this way: your future self will thank you for being thoughtful and proactive!

Get the daily newsletter search marketers rely on.

Processing…Please wait.

SUBSCRIBE

See terms.

function getCookie(cname) {
let name = cname + “=”;
let decodedCookie = decodeURIComponent(document.cookie);
let ca = decodedCookie.split(‘;’);
for(let i = 0; i <ca.length; i++) {
let c = ca[i];
while (c.charAt(0) == ' ') {
c = c.substring(1);
}
if (c.indexOf(name) == 0) {
return c.substring(name.length, c.length);
}
}
return "";
}
document.getElementById('munchkinCookieInline').value = getCookie('_mkto_trk');

What happens when marketers do not build new measurement frameworks?

For over two decades, marketers have utilized third-party publisher cookies to track their media performance. This method isn’t perfect, but it’s been a standard practice that’s set to evolve in a major way during the next 12-15 months.

From a digital marketing perspective, one of the most significant impacts is the loss of conversion measurement. This loss of performance data includes sales, sign-ups, purchases, revenue and other engagement metrics since those actions are likely to be restricted.

If marketers do not evolve their measurement practices, their accounts will rely on algorithmically-driven modeled conversions. 

Successfully enabling automation within PPC is critical to driving positive results.

One of the most potent algorithmic elements is smart bidding. Algorithms that drive cost-per-acquisition (CPA) and return-on-ad-spend (ROAS) bidding need strong data signals to optimize performance.

The data that feeds these algorithms must be reliable so that accounts are optimized toward the most valuable actions and this conversion data needs to have enough volume to drive machine learning.

Data loss means bid algorithms will not function properly, which will result in decreased PPC performance. Let’s try to avoid this!

More conversions will be algorithmically modeled as a result of signal loss

There is too much at stake (i.e., money) for ad platforms such as Google and Microsoft to leave marketers without another option to gain back lost data.

When marketers forge new measurement frameworks via Enhanced Conversions (EC), Google Analytics 4 or Offline Conversion Tracking, those are considered Observed Conversions.

This mix of first-party data and user-matched data (EC) is generated by registered actions taken by our website visitors.

Try to collect as much observed conversion data as possible.

The alternative is Modeled Conversions in Google and Smart Goals in Microsoft Ads. According to Google, Modeled Conversions is:

“When Google surfaces modeled conversions in Google Ads, we are predicting attributed conversions. In most cases, Google will receive ad interactions and online conversions but is missing the linkage between the two. The modeling we perform is modeling whether a Google ad interaction led to the online conversion, not whether a conversion happened or not.” 

Even after these large-scale privacy shifts, Google will continue to acquire mountains of data per user: search history, browsing history, and any other online activity when someone is logged into their Google Account, especially when those signed-in users are on a Google property.

Google will not be able to install a tracking pixel for that user specifically, but they should have enough data to algorithmically predict which media interactions lead to a conversion for an advertiser. 

Microsoft Ads is working on a version of conversion modeling. This product is called Smart Goals.

According to Microsoft:

“Smart Goals use Microsoft Advertising machine learning models to identify the best sessions on your website. If you have the UET tag set up correctly, the smart goal will examine all your website sessions and determine which of those sessions can be considered a ‘conversion.’ Smart goals use multiple signals to identify conversions. Some of the signals that are used include session duration, pages per session, location, device and browser.”

In essence, they are similar to Google’s modeled conversions. They both rely on machine learning at scale to understand user behavior and potential reactions to paid media exposure.

Marketers need to provide numerous additional signals to make any modeled conversions as accurate as possible.

With the loss of user-level data, modeled conversions will be part of the measurement landscape going into 2023.

This brings us back to creating a strong framework for supplying as much Observed Conversion data within the platforms, which will help inform the Modeled Conversion algorithms. 

Marketers have time and tactics to forge new measurement frameworks

The prospect of rebuilding your measurement framework can feel daunting, but you have the next couple of quarters to determine which solutions work best for you and your business.

Now is the time to start evaluating your current processes, review the new measurement tactics that are currently available and begin building a plan. 

In my last article, we laid the groundwork for what this metamorphosis means for the digital marketing landscape and approximately when it should occur. This article has addressed why adapting to these changes needs to be a strategic priority.

Next time, we can begin drafting a plan on how you can build a privacy-centric measurement and audience framework for 2023. 

The post How your PPC conversions will be impacted without privacy-first measurement appeared first on Search Engine Land.

Meta rumored to develop “Basic Ads” in response to recent privacy changes

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on Meta rumored to develop “Basic Ads” in response to recent privacy changes

Last week it was reported that Meta Platforms’ Facebook is in the early stages of planning a Basic Ads product. Basic Ads would be aimed at advertisers who are looking to build awareness around their brands. 

What are Basic Ads?

With the release of iOS14 in 2020, and the option for users to opt-out to having their data collected, Facebook advertisers are having a more difficult time reporting on performance. A Basic Ads product would offer and report only the simplest metrics such as engagement and video views. 

The initial response

Since Meta hasn’t officially announced the new product, little has been said about it in the advertising community. But Facebook veteran Curt Maly of Black Box Social Media who is aware of Metas plans had this to say:

“It’s interesting to think how a basic product with no targeting and no objectives would be beneficial. 90%+ of all the online marketers I know are focused on direct response where efforts can be directly tracked… with this basic ads platform, it appears that tracking will be rather difficult. 

Branding and awareness are used by larger companies with deeper pockets, most small businesses can’t compete with a brand who spends money “branding,” small business owners need/want to drive results now.”

He adds, “Brand new online advertisers and ad vets flock to FB to see fast results from targeting, conversion objectives and tracking results. If these 3 major goal lines are moved, I think we will be seeing a lot more people flock to Google/YouTube ads, TikTok ads and Apple’s new ad platform. 

I mean Apple didn’t update iOS to “help protect users”, Apple collects all the info they block on FB, Apple is about to get into the ad platform game once again and this is yet another reason for people to flock to a BETTER ad platform.”

Privacy changes effect on Meta revenue

Meta estimates a $12.8 billion hit to their revenue in 2022 due to the Apple changes. Meta CEO Mark Zuckerberg also reports “unprecedented levels of competition” from new platforms such as TikTok. In April, Meta laid out a three-tier plan to save it’s ad business in the wake of Apples privacy crackdown. Meta COO Sheryl Sandberg describes the strategy as “doing more with less data.”

Meta’s response

Facebook declined to comment on the new product. There has not been a release date for a rollout, though Business Insider reports testing to start in the EU ahead of the US.

What this could mean for advertisers

The Facebook ads platform was previously known for its wide range of targeted audiences and demographic options. As of late it seems like a lot of advertisers and businesses are moving away from the platform, reporting a decline in performance. Basic Ads could be a good alternative if they’re able to follow GDPR regulations while still providing useful targeting options. Basic Ads may work well for household names such as Nike or Netflix whose goals are engagement and awareness. But smaller businesses that rely on more granular targeting and lead generation, such as courses for business owners, or youth soccer camp registrations, may have a more difficult time and may abandon Facebook for good. 

The post Meta rumored to develop “Basic Ads” in response to recent privacy changes appeared first on Search Engine Land.

Switching from Google Analytics—here’s what you need to know

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on Switching from Google Analytics—here’s what you need to know

Switching from Google Analytics—here’s what you need to know

Google has dominated the data analytics marketplace for so long that it might feel daunting to switch. You may have reservations about an alternate platform because you have been conditioned to think there is no other tool that can live up to Universal Analytics.

We are here to reassure marketers that is not the case—switching has never been easier. Especially since Google recently announced they would be sunsetting their Universal Analytics platform.

If you made it here, you are likely to have already done your research on the pitfalls and limitations of Universal Analytics’ successor, Google Analytics 4. If not, it is worth reading ‘Google Analytics 4: drawbacks and limitations—is it worth sticking around?’.

Chances are, you have taken notice of the laundry list of issues with Google Analytics 4 and are considering jumping ship. This could be because of data quality or limitations, a lack of privacy-friendly features, or transparency in handling data.

As a marketer or analyst, you need to make an informed decision and choose the platform whose feature sets fit your organization’s needs to process user-level data while building trust with visitors. This article will help relieve some of the cognitive dissonances you may be experiencing. Let’s get into it:

Which is better, a free plan or a paid one?

It all comes down to the compromises and limitations you can accept depending on your project. Here are some key differences:

Free plan

Free web analytics platforms always have constraints—whether it is reduced performance, lack of solid privacy features, or sampled data. Nevertheless, a free web analytics platform can help you run your projects successfully.

Certain platforms ensure an easy upgrade from free to paid and more advanced versions. They also let you keep the setup, data, tags, etc., so you do not have to start collecting data from scratch. Others just eliminate the limitations and lift the price.

Paid alternatives

Paid alternatives overcome the limits of their free forerunners and grant more options for data collection and analysis. This includes the performance of the analytics platform, data protection, and extensive customer support.

For many organizations, the standard data safeguards of public clouds are sufficient. However, those operating in jurisdictions with strict data privacy regulations or highly regulated industries, like finance or healthcare, might require a private cloud or on-premises hosting. This is when having a paid plan would be necessary.

Piwik PRO offers both a free and paid plan so different kinds of organizations can get the service tailored to their needs.

Which is better, public, private cloud, or a self-host cloud?

Storing data on in-house servers used to be the gold standard for executing analytics safely and securely. Though, cloud-based analytics platforms have been steadily gaining market share.

Here is how they differ:

Public cloud

Public cloud is the most widely used model on the market. Storage space is shared between all the organizations that lease the environment from the same company. Public clouds are recommended for small and medium-sized organizations since they are more cost-efficient and quicker to deploy than the other options.

A drawback of public clouds is storing data on the external provider’s servers. The responsibility for data security rests with the provider. While this may be convenient, it limits how you can protect your data.

Some companies need to comply with strict internal data protocols or data residency regulations. With Piwik PRO, marketers can choose between four cloud hosting options— Germany, the Netherlands, the U.S., and Singapore.

Private cloud

Another type of deployment is through a private cloud. This model is often called “company cloud” or “organization cloud” because it is built exclusively for one organization. Companies that want to, or must separate their data from other cloud users who choose private cloud. This gives them the convenience of using the a public cloud with better data security.

A downside of a private cloud is the cost is significantly higher than a public because it requires the building of new infrastructure. Additionally, the installation time for private cloud is typically longer than with public cloud. For example, with Piwik PRO, the technical onboarding for private cloud clients takes about two weeks.

On-premises (self-hosted)

In an on-premises model, the infrastructure is built and managed by the company or organization, with employees having physical access to all resources. The company stores the data on its own servers or servers that they rent or lease from third parties.

Storing data in a chosen location and having full physical access to the infrastructure allows companies to process personal data securely. However, the costs related to the on-premises solution are significant and building and running your own data center is a long and complicated process.

Myths vs. Facts about switching analytics platforms

Despite the advancements of alternative platforms, some organizations are hesitant to make the move away from Google Analytics. To help dispel common fears and bring awareness to other platforms, here are the common misconceptions about switching:

I’ll lose historical data

This is a fact, but not for long. Some alternatives have developed data importers in the wake of Universal Analytics being deprecated. 

Alternatives are not integrated with Google

This is a myth. Alternatives are designed to work seamlessly with other Google solutions and partner products.

I miss some reporting capabilities

This is false. Each alternative has unique reporting capabilities, and some are very flexible, allowing for more transformations and data exports than Universal Analytics.


It is easier to run advertising campaigns with Universal Analytics

This is true. There is deep integration with Google Analytics and Google Ads/Google Marketing Platform, which gives access to an extensive repertoire of data.

I’ll lose my rank in Google Search

This is a myth. Alternatives’ customers do not report a lower rank in Google Search. Make sure your site is fast, mobile-friendly, popular (links), and with complete metadata.

Where should you go from here?

As the importance of compliance with regulations is growing, marketers should consider analytics platforms that offer a full set of privacy features.

Our intention with Piwik PRO Analytics Suite has always been to give clients powerful analytics capabilities along with key privacy and security features.

While we cannot tell you exactly what platform is right for your organization, we can help with your research.

If you’d like to learn more about Google Analytics alternatives or get more information on the Piwik PRO Analytics Suite, visit piwik.pro.

The post Switching from Google Analytics—here’s what you need to know appeared first on Search Engine Watch.

Google Analytics 4: drawbacks and limitations—is it worth sticking around?

Posted by on Jun 7, 2022 in SEO Articles | Comments Off on Google Analytics 4: drawbacks and limitations—is it worth sticking around?

Google Analytics 4: drawbacks and limitations—is it worth sticking around?

The free version of Google Analytics, Universal Analytics, is the most widely used web analytics solution. The platform is so popular that it dominates 86% market share, making Google the market leader. But even though many consider Google Analytics the standard, there are reasons to ask if it is the perfect choice for your marketing setup—especially since Google announced the sunset of Universal Analytics.

On July 1, 2023, Universal Analytics properties will stop processing new hits, forcing users to switch to its successor, Google Analytics 4. While this may seem like a natural progression, marketers should not be fooled.

The learning curve will be steep—Google Analytics 4 is almost an entirely new platform and still developing. On top of that, Google Analytics risks losing the users’ trust due to gray areas around the likes of privacy and data ownership.

With a privacy-focused future ahead, now is the time to seek alternatives that better balance data collection with compliance. With a proper analytics platform, marketers make your data collection as it should be: predictable and sustainable. After all, marketers and analysts want to process user-level data while building trust with their visitors.

In this article, we will dive into the issues with Google Analytics 4 from a user perspective and from a privacy and compliance standpoint, so you can make an informed decision before switching platforms.

User perspective: Google Analytics 4 is a step in the wrong direction

 Google Analytics 4 introduces an altered reporting and measurement technology that is neither well understood nor widely accepted by the marketing community.

From a user experience perspective, many find GA4 challenging to navigate. But beyond that, there are a host of challenges with the feature sets. Let us dig deeper into those limitations: 

There is no simple way to migrate your data

 Migration is a complex process and should be planned carefully. Unfortunately, Google Analytics 4 does not make it any easier. Without data or tag migration, all historical data from Universal Analytics will not be transferred to the new platform.

The challenge only grows with the organization’s size—you can have hundreds of tags to move. So, if marketers must start collecting data from scratch, they might as well switch to a new analytic software.


Not-so-intuitive user interface

The most prominent challenge marketers and analysts will likely encounter with Google Analytics 4 is the unfamiliarity with the new interface.

A new dashboard has several immediately apparent differences from what marketers are used to operating. Hit types are essential to how Universal Analytic properties handle all stats. Hit types include page hits, event hits, eCommerce hits, and social interaction hits.

GA4 doesn’t have any concept of a hit type like Universal Analytics uses. Everything in Google Analytics 4 is categorized as an “event.” This is a huge difference.

In order for marketers to have success on the new platform, they will have to adapt quickly to maintain the same momentum they had with this previous platform.

Limits on custom dimensions

A custom dimension is an attribute that marketers can configure in their analytics tool to dive deeper into their data. It gives the option to pivot or segment this data to isolate a specific audience or traffic for deeper analysis.

GA4 indeed allows for custom dimensions to segment reports, but there is a strict limit. You can only have up to 25 user-scoped custom dimensions and up to 50 event-scoped custom dimensions per property.

Lack of custom channel grouping

Channel groupings are rule-based groupings of marketing channels. When customized, these groupings allow marketers to track the performance of those channels efficiently.

Unlike Universal Analytics, GA4 does not allow you to create custom channel groupings in the new interface. Instead, marketers will only be able to use their default channel groupings.

Motivations behind the short deadline

 The deadline Google has left the analytics community to act is startling. There are many speculations as to why this might be, including:

Google may have been disappointed with the speed of adoption for Google Analytics 4 and decided to act decisively.
Google circumventing some of the legal heat that Universal Analytics is facing in the EU.
Google wanting to cut costs and rid itself of technical debt associated with thousands of websites with legacy solutions installed. Since GA4 is designed to support Google’s advertising network, it guarantees more revenue than the competition.

Now there is a concrete deadline to make the switch, marketers will need to decide whether they want to start adjusting to Google Analytics 4 or start afresh with a new platform.

Privacy and compliance: Google Analytics 4 has a long way to go

If a company operates in multiple countries, marketing teams will need to be aware of the numerous challenges resulting from the obligations of both local data privacy laws and international regulations.

Data protection legislation constantly changing and tight security regulations only complicate things further. Reading the tea leaves, we believe GA4 will not last long in Europe. Here’s why:

Google Analytics violates European law

Google makes it difficult to collect data in line with the General Data Protection Regulation (GDPR), which aims to restore control of personal data to users and customers. The regulation requires you to obtain explicit consent when processing personal data. Failure to comply with this provision can result in hefty fines or even prosecution.

The recent decision of the Austrian Data Protection Authority (DSB), states that the use of Google Analytics constitutes a violation of GDPR. This means that organizations engaged in gathering, storing, and processing data about EU citizens have to adjust their policies and introduce serious technological changes to be GDPR-compliant.

There is no clear guideline where the data is connected through Google Analytics

A Google guide implies data is transferred to the closest Google Analytics server hub. However, the data may be stored in a geographic location that does not have adequate privacy protection to the EU.

Newly introduced features in GA4 partially address this concern by allowing the first part of data collection (and anonymization) on European servers. However, data can, and most likely will, be sent to the U.S.

 The future of marketing requires users’ consent

Whether it be the data quality, tool limitations, lack of privacy-friendly features, or transparency in handling data, we believe marketers will likely consider switching platforms.

Piwik PRO excluds the privacy and compliance issues associated with Google Analytics, allowing marketers to collect data predictably and sustainably. The user interface and feature sets are similar to Universal Analytics, so marketers and analysts feel at home when switching to our platform.

If you would like to learn more about Google Analytics alternatives or get more information on the Piwik PRO Analytics Suite, visit piwik.pro.

Still undecided? Check out our article on addressing the concerns about switching to an alternative analytics solution and the analytic mindset you should be taking: Switching from Google Analytics—here’s what you need to know.

The post Google Analytics 4: drawbacks and limitations—is it worth sticking around? appeared first on Search Engine Watch.

How One Simple Tool Will Cut Down Your Blogging Time by Half

Posted by on Jun 6, 2022 in SEO Articles | Comments Off on How One Simple Tool Will Cut Down Your Blogging Time by Half

How One Simple Tool Will Cut Down Your Blogging Time by Half

You heard about AI, but have you ever used it?

Just think of it this way, what if you can use artificial intelligence to create content for you so you wouldn’t have to even write?

Well, believe it or not, this blog post that you are reading now was written by a simple AI tool.

To be fair no AI tool is perfect yet as technology isn’t there yet.

And to give you perspective the AI wrote this post and then I had to modify it and add specific sections and steps.

When I look back and did an analysis, AI wrote roughly 25% of this post.

That’s not too bad. Sure, I had to write 75%, but the hardest part about writing isn’t putting words on a paper it’s coming up with the ideas and the overall format.

So today I am going to show you how to write a blog post using AI… keep in mind over time it will get better as technology improves.

But first, something you should know…

Don’t worry about duplicate content

Google doesn’t penalize for duplicate content and even if they did this AI tools makes 100% unique content for you.

So, you won’t have duplicative or plagiarism issues.

It’s not just mixing words around it is trying to understand what you are trying to write your article on, do research, and come up with words that encompass everything you are trying to get across.

Let’s get started

First off, go to Ubersuggest’s AI Writer.

First, enter in the keyword you want an article on. For this example, I chose the term “digital marketing” as my blog is on digital marketing.

Then you will want to select a title from the list provided. You don’t always have to use it at the very end but a lot of the titles are based on what people are searching for.

Next, you will want to choose a meta description.

Similar to the above, meta description is also based on keywords that are popular from a traffic perspective.

After you select your meta description you will want to select a few headings.

Keep in mind headings are like book chapters. Pick the ones that work the best, and again keep in mind you can always adjust the text later on.

The last step is, that Ubersuggest will take all of the data and spit out content for you.

By no means is the content perfect and I wouldn’t recommend publishing it or any AI content written by any tool for that matter.

More so use it as a starting ground.

So how do you modify the article?

Well, the AI tool won’t always be contextually accurate, but most of the time it is.

More so the way you make the AI piece amazing is by getting more detailed.

So for example in the digital marketing above, in the “create an online presence” section I would go in-depth on Facebook, Twitter, LinkedIn, and Pinterest on how you can use them for marketing.

As for Google + it doesn’t exist, so I would delete that part and expand the other platforms and give step-by-step instructions on how to leverage them.

I would also include screenshots so each step is clear.

Under the “promote your brand” section I would discuss ads and other ways to promote your brand such as running promotions, sales during holidays, or even doing webinars.

I would give examples of other successful ones that businesses have run and give examples of failed campaigns that have been run and break down what you can learn from them.

The “grow your audience” section already talks about what to do if you have no followers. They are all good tips that the AI tool wrote, but the tips are too surface level. I would dive deep into each one.

From how you should think about commenting on other profiles and blogs to what email templates to use to convince people to collaborate with you.

And the measuring success section was cut off from the above screenshot but it mentions Google Analytics. I would go more in-depth on how to set up goals in Google Analytics and tie in revenue.

Conclusion

The Ubersuggest AI tool isn’t meant to replace humans or for it to perfectly write your content for you.

It’s meant to give you a start.

You have experienced writer’s block and we all hate it. The tool solves that part plus it gives you a head start.

It not only gives you title and meta description ideas, but it writes some of the content for you as well as gives you popular headings based on keyword search volume.

All you have to do is adjust and fill in the rest.

So make sure you give it a try.

Free webinar: Get a jumpstart with Google Analytics 4

Posted by on Jun 6, 2022 in SEO Articles | Comments Off on Free webinar: Get a jumpstart with Google Analytics 4

Free webinar: Get a jumpstart with Google Analytics 4

Ready or not, Universal Analytics is sunsetting next year. While there is still some time to prepare, marketers should start moving over to Google Analytics 4 now.

Because GA4 operates across platforms and is designed with privacy at its core, it can provide a better user experience for marketers. But there is a learning curve involved in understanding the stark differences between GA4’s and UA’s interfaces. Join experts from iQuanti and learn how you can get started.

Register today for “Get a Jumpstart with Google Analytics Today!” presented by iQuanti.

The post Free webinar: Get a jumpstart with Google Analytics 4 appeared first on Search Engine Land.