A first-time visitor to Baidu could be forgiven for thinking they had wandered into Silicon Valley.

The Chinese internet company’s headquarters in Beijing, housing its 8,000-plus staff, resembles the spaces set up by its Californian counterparts to inspire employees to dream up world-changing ideas. The massive glass building comprises wide open spaces with natural light, a baby nursing room, and even roof gardens – where one young man is napping in the midday heat.

But the reality of working at the company is rather different from this first impression. In a third-floor meeting room, the group of young engineers gathered around a table cluttered with laptops and crumpled junk food wrappers, are not paid to dream.

At Baidu, the priority is not in producing cutting-edge technology but ultimate functionality, explain company executives.

Wang Mengqiu, senior director of technology and products, says Baidu’s product development philosophy differs from [rival search company] Google’s focus on “very cool” technology. “Our logic is different – we think about what users need most,” she says.

Ms Wang is a Baidu veteran, having joined in 2003 after obtaining a masters from UCLA and working at a San Francisco-based start-up for a year.

The pig-tailed executive’s insights provide a rare glimpse into the culture of a traditionally reticent company.

Baidu has been reluctant to open up, particularly to foreign media, because it has been stung by portrayals in the media that it is a Google imitator.

Ms Wang is defiant in the face of such suggestions. “I don’t care that many people say Baidu can’t innovate,” she says. “You have to ask whether completely new things are needed.”

She says Baidu would never have developed a product like Google Earth, for example. For China’s nearly 500m internet users – Baidu’s target market – Google’s interactive world map has very little value, she argues.

“It is a dazzling, very cool product, but really think for a moment. The users we need to consider are not just high-end, well-educated users,” she says.

Baidu’s ideal of a useful product is Baidu Post Bar, a forum that combines search and social networking to help internet users link up with people with the same needs and interests based on the web search keywords they use.

Its focus has also been on features that make web search more suitable to Chinese tastes. The size of its search box, for example, is a better fit for average phrases in Chinese characters. And the company was an early provider of predictive search, a function especially important in China, as typing is more cumbersome than in alphabet-based languages.

On the other hand, Baidu has refrained from spending too much time and money researching how to search the content of pictures and videos.

To reflect its different priorities, Baidu’s product development is driven not by engineers, as at Google, but by product managers.

They conduct extensive research among the search engine’s users to define product profiles. Research engineers, who enjoy almost cult status in many western technology firms, are relegated to finding technical solutions for what the product managers want.

To make the whole process as down-to-earth as possible, the product marketing department, which wields the most power in the process, is staffed with people with practical experience among users and customers of web search.

Some are internet boffins – people who spend huge amounts of time online and have a thorough understanding of applications and internet user habits. Others have founded or run websites, worked as editors at news portals, or are degree-qualified information management graduates.