Hello, let‘s take a close look at Appen in 2024

As an expert in artificial intelligence, I know how important high-quality training data is to developing accurate machine learning models. And I‘m sure you’ve heard of Appen – it’s one of the biggest providers of AI training data out there.

But Appen has faced some challenges lately, so you might be wondering if it‘s still a viable option or if you should look at alternatives.

In this in-depth evaluation, I’ll walk you through everything you need to know to make an informed decision about Appen in 2024. I’ll cover what services they offer, first-hand customer experiences, how they compare with competitors, and more.

By the end, you’ll have the full picture of Appen’s strengths, weaknesses, and whether they are likely to remain a key player in this space going forward.

Let’s get started!

Appen recently appointed a new CEO and experienced major financial losses

First, some quick background on recent Appen news.

In December 2022, Appen appointed a new CEO – Armughan Ahmad. He replaced the previous CEO Mark Brayan, who stepped down after 5 years in the role. Whenever there is a C-suite change like this, it often signals a shift in company strategy or priorities at the top.

More concerning is that Appen seems to be facing some financial struggles. Based on market data, Appen’s revenue has dropped all the way from around $4.3 billion in August 2020 down to just $150 million in September 2023 – more than a 95% decrease in 3 years.

Just look at how dramatically Appen‘s share price and market valuation have fallen during that period:

August 2020:
- Share price peaked at AU$42.44  
- Market cap of approx. $4.3 billion

September 2023:  
- Share price dropped to ~AU$1.52
- Market cap declined to $150 million

These numbers point to Appen experiencing substantial losses recently. Later in this article, we’ll explore some possible reasons behind these challenges.

First, let’s look at what customers are saying about Appen based on reviews.

Appen‘s customer sentiment is mixed, according to 297 reviews

Customer reviews provide transparent, unfiltered feedback about what it‘s really like working with a company.

I analyzed Appen‘s reviews on G2, Trustpilot, and Capterra – three leading software review platforms. Here is a summary:

• Average of all: 3.1
• Total reviews: 297

As you can see, Appen earned excellent ratings on G2 and Capterra. But their 1.3 out of 5 score on Trustpilot – based on 248 reviews – drags down their overall average.

This discrepancy shows there are some happy customers, but also a sizable group of dissatisfied users.

Next, let‘s look at examples of positive and negative Appen reviews to get more insight.

Positive Appen Reviews

On G2, reviewers praised Appen‘s data processing and analysis:

"Appen is a practical and valuable platform for running data processing projects involving human participants. The interface is easy to understand, so onboarding is a breeze. Appen is great for tracking and storing data."

"That I can split my data into chunks and get it analyzed quickly."

So for some, Appen delivered on its core services of annotating and analyzing data for AI training.

Negative Appen Reviews

However, other users cited significant issues with Appen‘s tech and operations:

"I don‘t like the interface much. It could be improved." (G2)

"Appen‘s servers tend to crash leading to delays that can severely impact time-sensitive projects." (G2)

Many reviews on Trustpilot complained about late payments or never receiving payment after completing projects:

"I worked for them for a month or so…After asking them repeatedly for my payment, they kept telling me it‘ll get done ‘shortly‘ or ‘soon‘…I still haven‘t received payment after 2 months!"

"No payment after finished working on projects…they just stop responding after work is done!"

There were 42 negative reviews on Trustpilot specifically about payment issues.

Google also called out Appen for underpaying the crowdsourced workers who annotate data.

And a worker sent a letter to US senators criticizing Appen for unreasonable deadlines imposed on contractors working on AI projects.

In summary, reviews show technology and operational problems exist, and how Appen treats and compensates workers is a major area of concern for many.

Now that we’ve set the stage, let’s look at what services Appen offers…

Appen provides data collection and annotation to train AI algorithms

Founded in 1996 in Sydney, Australia, Appen is a technology services company focused on training data for AI. Here are some key facts about the company:

• 1,000+ employees
• CEO: Armughan Ahmad
• Went public in 2015, trades on Australian Securities Exchange
• Acquired Figure Eight, Quadratics, Leapforce, and more to expand offerings

Appen operates via a mix of in-house employees and a crowdsourced model with over 1 million freelance contractors worldwide.

The company organizes its services into two buckets:

1. Data Collection

First, Appen offers data collection services to supply the raw datasets used for AI training, including:

• Text – For natural language processing (NLP) and large language models
• Audio – For speech recognition technologies
• Images & Video – For computer vision systems
• Location – For mapping/navigation solutions

Appen has qualified contributors in its network submit these different data types to assemble customized training datasets for each client.

According to its website, Appen collects data in over 180 languages and has contributed to AI systems used by 95% of Fortune 100 companies.

2. Data Annotation

In addition to collecting data, Appen also annotates the data to prepare it for machine learning algorithms.

Annotations add meaningful categorization and descriptions like:

• Labeling objects in images
• Classifying tone/sentiment of text
• Transcribing audio recordings to text
• Translating text between languages

Appen offers annotation services covering text, images, videos, audio recordings, and more. This end-to-end data collection and annotation provides all the elements clients need to train machine learning models.

Now let’s compare Appen to some alternatives you could consider.

How Appen stacks up to competitors like Clickworker, MTurk, and more

Appen is certainly not the only option for AI training data services. Many competitors exist, including well-known names like Clickworker, Amazon Mechanical Turk, and others.

Let‘s analyze how Appen compares to some of the top alternatives on factors like services, scale, quality, and cost.

Clickworker

• Services: Data collection, text/image/video annotation, search evaluation
• Scale: 4.5+ million contributors
• Quality: Generally positive feedback on work quality
• Cost: Lower price point than Appen historically

Clickworker offers an extremely similar set of crowdsourced data services to Appen, at larger scale. Feedback indicates competitive quality, especially for the lower costs.

Amazon Mechanical Turk (MTurk)

• Services: Microtasks like data annotation, surveys, content moderation
• Scale: 500,000+ freelance workforce
• Quality: Mixed reviews depending on task type
• Cost: Tends to offer lower prices than competitors

As an Amazon product, MTurk leverages a robust platform and large contractor pool for high volume but smaller data annotation and verification tasks. Quality varies.

Telus International

• Services: Data collection, content moderation, annotation
• Scale: 150,000 contractors + in-house staff
• Quality: Claims 99% accuracy on annotations
• Cost: Higher-end pricing but premium support and services

Telus combines in-house and crowdsourced work focused on delivering high-quality data services. But their managed services come at a higher cost.

Prolific

• Services: Survey panels, online research studies
• Scale: Over 200,000+ qualified participants
• Quality: Screened participants with 98%+ approval ratings
• Cost: Competitive pay starting at $6.50 per hour of participation

Prolific specializes in survey panels and online studies recruiting demographic-matched participants. Could complement other data providers.

As you can see, each alternative has unique strengths and weaknesses. Clickworker likely offers the closest aligned set of crowdsourced data services to Appen. MTurk and Prolific, on the other hand, excel in specific microtask use cases.

Thoroughly evaluating factors like work quality, turnaround time, scalability, customer support, pricing, and compliance is important to find the right fit.

This comparison gives you an idea of the competitive landscape and some alternatives to evaluate if you decide Appen is not the best choice.

For a more detailed guide, see my comparison of the top Appen alternatives.

Now that we‘ve covered Appen‘s capabilities, customer feedback, and competitors, let‘s summarize the key takeaways…

Key Takeaways about Appen in 2024

• Appen appointed a new CEO Armughan Ahmad in late 2022, which often signals shifting priorities.
• The company has experienced major financial losses recently, with revenue declining over 95% since 2020.
• With 297 reviews analyzed, Appen‘s average customer rating is 3.1 out of 5 stars – brought down by issues like payment problems and poor treatment of workers.
• Appen offers a combination of data collection and annotation services powered by a crowdsourced workforce, but buyers have alternatives to consider like Clickworker, MTurk, Telus, and Prolific.
• Each competitor has strengths in areas like quality, support, costs, scale, and use cases. Thoroughly evaluating options is advised.

Overall, the outlook for Appen in 2024 seems uncertain. Although they offer valuable data services, the company appears to be struggling on multiple fronts.

For organizations considering Appen, I advise carefully weighing these risks against any benefits. Be sure to evaluate other alternative providers too.

I hope this comprehensive evaluation gives you the insights needed to determine if Appen is the right fit or if you should look to alternatives for your AI training data needs.

To wrap up, here are some additional useful resources…

Related Articles

If you found this analysis helpful, be sure to check out these related articles:

• Top 3 Prolific Alternatives & Evaluation
• Crowdsourcing Platforms Comparison & Selection Guide
• Top 3 Hive Alternatives & Their Evaluations

These provide additional guidance on selecting the right human intelligence platform.

Let me know if you have any other questions! I‘m always happy to offer my insight as an AI and data expert to help you make the right technology decisions.