Target, Adobe Target, CRO

– Martech

Warning: The tone of this article may sound aggressive to some audiences. It’s not like I am heavily biased to a particular side. Indeed, there is no offence intended for anyone. It’s just - I am such an arse whose personality is quite raw, and I don’t give a F on tone polishing anymore. Sounds unprofessional? Maybe. But this is my messy brain garden.

What is CRO and Why It’s Important in Web Development

Conversion Rate Optimisation (CRO) is a systematic process of increasing the percentage of website visitors who take a desired action—whether it’s filling out a form, becoming customers, or otherwise. In the competitive landscape of web development, CRO is crucial because it directly impacts the bottom line. A well-optimised site can significantly boost revenue without the need for additional traffic, making it a cost-effective strategy for businesses.

The Frustrations of CRO Campaigns

Hush, little developer. Don’t you cry…

Why do CRO campaigns often lead to head-splitting situations? When working on A/B testing for clients, there are frequently maddening occurrences. Let’s briefly touch on CRO MarTech tools first. There are various tools like Google Optimize, Optimizely, Adobe Target, and so on. They all boast about how robust and powerful their tools are, and countless marketers and agency account managers believe that with a developer who can handle code appropriately and a powerful tool, there’s no A/B testing they can’t accomplish.

But I strongly disagree with this notion.

The Real Challenges

In my over ten years of working in the frontend industry, what slowed down development speed and productivity wasn’t my lack of self-improvement or proficiency with tools.

  1. Communication Breakdowns: Internally, it’s the failure to properly convey the developer’s Yes and No to the client, and the unresolved agreements in communication with the person in charge who has to deal with the client.
  2. Unrealistic Expectations: Externally, it’s because the client can’t accept the problematic issues or implementation limitations (or rather, the lack of cooperation in understanding the client’s system) that were concluded and keeps pestering the developer until they get what they want.

Problematic CRO Requests and Their Technical Implications

One of the most maddening CRO requests I’ve encountered involved a client who wanted to test multiple landing page variations simultaneously. While this might sound straightforward, the technical implications were anything but. Each variation required different scripts and styles, leading to conflicts that caused the site to break in unexpected ways. The client’s insistence on using a third-party A/B testing tool only compounded the issue, as it injected code that was incompatible with our existing framework.

The Ever-Shifting Maze

You navigate this maze, desperately trying to find an exit, only to realise that the walls keep shifting. One moment, you’re implementing a simple button colour change; the next, you’re expected to completely overhaul the site’s architecture because “the client saw something cool on a competitor’s website.” It’s a transformative experience, really. You enter as a bright-eyed developer and exit as a hollow-eyed husk.

Let’s dive deeper into the hellish landscape of CRO campaigns, shall we? First of all - account managers. Why are we trying to implement a client’s fever dream of an A/B test?

The boulder? It’s made of conflicting requirements, unrealistic expectations, and a complete disregard for the laws of physics—or, in our case, front-end DOM development.

You push and push, explaining why their desired changes will not happen smoothly. Worse, sometimes our small piece of contained code is delivered through A/B code injecting platforms (in this case, Adobe Target—but once again, this is not about the Target Problem. It is from lacking a ‘How DOM works properly’ problem.)

However, what can you do? You are just the lowly developer trapped in a labyrinth of nonsensical requests and impossible deadlines.

And let’s not forget our beloved managers, those visionaries who can see everything except the blood, sweat, and tears of their dev team. They waltz into meetings, spewing buzzwords like ‘synergy’ and ‘disruptive innovation’ while we’re left wondering if they’ve ever actually seen a line of code. ‘Just sprinkle some AI on it,’ they say as if artificial intelligence is some kind of digital fairy dust. Sure, boss, let me just call up my good friend Skynet and see if it’s free to help out with our banner placement issues.

Long-Term Consequences for Developers and Businesses

Later, the developer ends up working overtime, forcibly pushing through something that can’t be done while listening to the angry complaints of account managers who have become equally emotionally drained from dealing with furious clients.

Even if a solution is somehow found, later, it’s common to hear from company directors or team leaders, ‘What? You spent two days just to change a cart ribbon?’

That’s what I’m saying, Dawg. In the first place, on what basis do they demand such unnecessary perfection when building up requirements or ideas? How much will customer inflow increase just by fixing the position of one banner?

Neither the client’s marketer nor the account manager can provide any evidence. I don’t know why the fuck I’m doing this trivial task of moving a banner from below the header to the middle of the landing page and why I have to listen to complaints about fixing it all because it looks awkward at a very specific point when accessing via mobile and tablet.

And let’s not forget the cherry on top of this shit sundae: the deadline. “We need this life by yesterday,” they say as if we developers have a secret time machine stashed away next to our ergonomic keyboards. Maybe I should start billing in theoretical physics hours instead of actual development time.

All while trying to maintain some semblance of code integrity and not succumb to the temptation of replacing our entire codebase with a single line: console.log("I quit").

Even thinking about it now, this is why I don’t want to get involved in CRO campaign projects again. Of course, it depends on who you work with, but at least not working with arrogant people who insist they’re always right and treat developers who can’t deliver as idiots is better for mental health.

 These CRO challenges are far-reaching. For developers, the constant pressure to meet unrealistic demands can lead to burnout and job dissatisfaction. This, in turn, affects the quality of work and increases turnover rates. For businesses, the failure to properly implement CRO strategies can result in wasted resources and missed opportunities. Instead of reaping the benefits of optimised conversion rates, companies find themselves stuck in a cycle of endless revisions and mounting frustration.

Oh…Shit. Here it comes. The damn AI Hype.

Unrealistic Promises and the Reality of Working with AI Tools

You think you’ve seen it all, and then comes a request to “just sprinkle some AI” on the website. Sure, let me just summon the ghost of Alan Turing and have him whip up a sentient chatbot that can predict user behaviour better than a crystal ball on steroids. Oh, and while we’re at it, why don’t we also make the website load faster than the speed of light? I’m sure Einstein won’t mind if we bend a few laws of physics in the name of conversion optimisation.

But who knows? Or maybe, just maybe, I’ll find that magical AI fairy dust that can turn these ridiculous demands into a functioning, high-converting website.

So far, I haven’t had much luck, though. I want to ask the CEOs who are selling their AI services, claiming that AI will solve everything and replace many developers and professionals. As a current master’s student in data science and a developer with ten years of experience who has been using Claude, ChatGPT, and Bing for the past two years, none of these have automatically generated perfect code in one go. Of course, I’m not saying these services are inefficient. They’ve done a good job as fast and efficient Rubber Duckies. But that’s it. In the end, to find accurate and appropriate solutions, I had to do indexed reinforcement training for prompts in my account’s internal index for each situation, and in that process, debugging and supervised training using my expertise were necessary. Even I couldn’t create such magical prompts, but countless startup founders on various social accounts and YouTube show demo videos of creating perfect prototypes by auto-generating designs and code along with AI’s perfection, but not a single person has revealed the detailed prompts and training process end-to-end on how they created those results. Surprisingly, no one talking about the AI hype right now is an operator actually staging real products with that AI or a developer with years of experience in machine learning algorithms. Does everyone except me know some secret? Or perhaps they are delusional or lying to sell their products or branding?

The Way Forward

Bridging the Gap Between Expectations and Reality

Well, my point is, even though you wish the AI use would boost the delivery speed and hope to replace the grumpy developers with it - it won’t solve the delayed delivery. The main cause was not from the developer. It’s from human interactions and psychology games.

Internally, it’s the failure to properly convey the developer’s Yes and No to the client and the unresolved agreements in communication with the person in charge who has to deal with the client. Externally, it’s because the client can’t accept the problematic issues or implementation limitations (or rather, the lack of cooperation in understanding the client’s system) that were concluded and keeps pestering the developer until they get what they want.

You need to understand how DOM works before idealising campaigns

But it’s quite a comprehensive subject. I know. Here is the simplest solution. Listen to what the front-end developer says without throwing disappointed eyes.

Front-end development and martech are closely related, and many martech problems can indeed be caused by incompetence in the front-end development sector. Front-end development plays a critical role in creating user interfaces and experiences for martech applications, and problems with front-end development can impact the overall effectiveness of MarTech solutions.

For example, if a MarTech application has a slow or poorly designed user interface, users may be less likely to engage with the application or may not use it as frequently as desired. This can result in a lower ROI for the martech investment.

Martech teams must work closely with front-end developers to ensure their user interfaces and experiences are optimised for performance, usability, and user engagement. This requires a deep understanding of both martech and front-end development and close collaboration between teams.

Ultimately, martech teams should prioritise front-end development as a critical component of their solutions and ensure they have the right expertise and resources to deliver high-quality, effective user experiences.

Obsession towards Anti-Flicker

Give me some suggestions for anti-flicker implementation relating to marketing tech.

Sure, here are a few suggestions for implementing anti-flicker technology in marketing:

  1. Use responsive design: One of the most common causes of flickering is when a website is not optimised for different screen sizes. Responsive design can help ensure your website looks consistent across different devices and reduce the likelihood of flickering.

  2. Implement lazy loading: When a website loads all its content at once, it can cause flickering. Implementing lazy loading, which loads content as the user scrolls down the page, can reduce the amount of content that needs to be loaded at once and prevent flickering.

  3. Optimize images: Large images can also cause flickering. Optimising your images by compressing them can reduce their size and prevent flickering.

  4. Use CSS animations: If you need to use animations on your website, consider using CSS animations instead of JavaScript. CSS animations are often smoother and can help reduce the likelihood of flickering.

  5. Test your website: Finally, it’s important to regularly test your website to identify any issues causing flickering. Use tools like Google PageSpeed Insights or GTmetrix to identify issues and make necessary optimisations.

Important Considerations Before Implementing Anti-Flicker Suggestions

Correct use of VEC

VEC stands for “Voice of the Customer” and is a marketing methodology to collect feedback from customers or website visitors. Here are some tips for the correct use of VEC:

  1. Clearly define your goals: Before using VEC, it’s important to clearly define your goals for collecting feedback. Are you looking to improve your website’s usability, gather ideas for new products or services, or address specific customer complaints? By defining your goals, you can tailor your VEC strategy accordingly.

  2. Choose the right channels: There are many different channels you can use to collect feedback, including surveys, feedback forms, customer reviews, and social media. Choose the channels that are most appropriate for your goals and audience.

  3. Ask the right questions: When collecting feedback, it’s important to ask the right questions. Avoid leading questions that may bias the responses, and focus on open-ended questions that encourage customers to share their honest opinions.

  4. Analyze the data: Once you have collected feedback, analyse the data to identify common themes and trends. Use this information to inform your marketing and business strategies.

  5. Take action: Finally, take action based on the feedback you have received. Use it to improve your website, products, or services and communicate the changes to your customers. This will help build trust and loyalty among your customers.

Give me some examples of the correct use of virtual experience composer from Adobe Target.

Sure, here are some examples of the correct use of Adobe Target’s Virtual Experience Composer (VEC):

  1. Testing different page layouts: You can use VEC to test different layouts to see which performs best. For example, you can test different homepage variations to see which drives the most engagement and conversions.

  2. Personalizing content: VEC allows you to personalise your website content based

Sure, here are some examples of the correct use of Adobe Target’s Virtual Experience Composer (VEC) in order to avoid flicker on the rendering page

Certainly, here are some examples of using Adobe Target’s Virtual Experience Composer (VEC) to avoid flicker on rendering pages:

  1. Implementing page hiding: VEC allows you to use page hiding to avoid flicker. This involves hiding the page content until it has fully loaded, so users don’t see any flicker or loading screens. You can use VEC to set up page hiding, so it is triggered when the page loads.

  2. Optimizing page load speed: Slow page load times can also cause flicker. You can use VEC to optimize page load speed by testing different variations of your website and identifying any issues that may be causing slow load times. For example, you can test different image sizes or server configurations to find the fastest-loading variation.

  3. Implementing lazy loading: VEC can also be used to implement lazy loading, which loads content as the user scrolls down the page. This can reduce the amount of content that needs to be loaded at once and prevent flicker.

  4. Testing different scripts: If you have multiple scripts on your website, they may conflict and cause flicker. VEC can be used to test different combinations of scripts to find the optimal setup that avoids flicker.

  5. Monitoring performance: Finally, VEC allows you to monitor your website’s performance in real-time, so you can identify any issues that may be causing flicker and make necessary adjustments. This can help you proactively address any issues and improve the overall user experience.

Related Articles

2022-08-17-CRO-Glossary-Target-Travelling to the office

2022-08-03-Website-Update-Jekyll-DrumConversion

2024-06-23-Things-need-know-about datalayer-tagging-in-react