Last week it unveiled the latest in a series of profit downgrades and its share price finished the week at A$2.87. Its market capitalisation – the amount share market investors value the company at – slide from $A5 billion to about A$3.5 billion.
On Thursday, the company downgraded its forecast operating earnings by more than 50 per cent.
Appen's tech company clients such as Google and Facebook owner Meta have cut back on investments in advertising-related AI projects.
Changes by Apple to its iOS platform have limited the customer tracking capabilities of digital advertisers, which has hit the advertising revenue of the tech giants and so they are putting less money into AI advertising.
It is now expecting operating earnings of just US$13 million to US$18 million.
Appen was founded in 1996 by Julie Vonwiller, a linguist at the University of Sydney, who started the company in her spare room.
She was later joined by her husband, Chris Vonwiller, who left his job at Telstra in 2000 to join full-time. He served as the company's chairman and the pair remain Appen's biggest shareholders.
Julie Vonwiller started by collecting speech data and preparing it for customers, who used the language data to help power applications such as chatbots and virtual assistants.
The business grew quickly and started categorising data on other things such as websites and search engine results.
Every time you use a search engine, look at social media or dictate a request to Apple's Siri, odds are that Appen data is helping to drive these applications.
Things looked good for the company, but the tech giants account for about 80 per cent of its revenue and once they cut back on their spending, Appen was in trouble.
One earnings downgrade is bad enough, but the October downgrade followed downgrades in May and August.
The rapid succession of downgrades, along with the company's admission that it was having difficulties providing guidance of future earnings, raises questions about how much visibility Appen has over its financial performance.
This is a big red flag for investors – if the company doesn't know what's going on with its financials, how can it take control of them and get them back on track? And how much confidence can investors have in any of its pronouncements?
At the start of August, for instance, chief executive Mark Brayan said tried to reassure the market after that month's earnings downgrade.
"The fundamentals of our business remain strong and our operational performance and the quality of our service we provide customers continues to improve, evidenced by higher [net promoter score]," he said.
"We are increasing our range of products and through our product investments remain well positioned to serve our customers. Despite the current challenging operating conditions, we remain committed to our longer-term growth strategy and confident of our prospects in the high growth AI market."
The shares closed at A$4.15 after that downgrade, but have slid a lot further since.
The company suffered another blow in June when Canadian IT company Telus walked away from a $9.50-per-share takeover bid.
The company didn't provide any explanation as to why it suddenly pulled the pin, leaving investors to wonder what it discovered.
The data sets which Appen provides to the tech companies might be used to power artificial intelligence, but they are created by human intelligence.
Appen has the tasks carried out by around a million workers around the globe to create its "human curated data sets".
These workers label data in small chunks of work for equally small payments. The company's business model is heavily dependent on these workers and that is a potential problem.
Many of these workers are in some of the world's poorest countries, and critics argue they are underpaid and that there is a huge power imbalance between them and Appen.
For instance, if a worker receives a bad review from a client, they can be suspended from working for Appen and left without income.
As investors put more focus on environmental, social and governance factors the tech giants are likely to respond by looking more closely at their supply chains, and this could include the employment practices of the data companies such as Appen.
Appen has a long hard road ahead of it.
