Microsoft has finally released its artificial intelligence-powered search engine, and some of the early reviews are good. But what will it mean in to the bottom line?
analyst call held Tuesday night gave some indication. First, the good news — Microsoft says for every point of share gain in the search advertising market, it’s a $2 billion revenue opportunity. According to data compiled by Statista, Google had 86% worldwide desktop market share of search in December, compared to 9% for Microsoft’s Bing.
The software giant made nearly $18 billion in advertising revenue last year, out of the roughly $200 billion of sales it brought in during the last calendar year.
Microsoft says it has a real opportunity to pick up market share outside the United States, given the language models offer better translation and a better understanding of local content. Microsoft pegs the total digital advertising market at $500 billion, with 40% of that coming from search.
But AI is expensive. “The new experience will be delivered at a higher cost to serve,” said Philippe Ockenden, chief financial officer for Windows, devices and search, according to transcript compiled by FactSet. But Microsoft CFO Amy Hood insists that all the new money coming in will be “incremental gross margin dollars for us, even at the cost to serve that we’re discussing.”
Hood was hopeful that costs will come down over time. “It’s a leveraged platform that starts at the supercomputer layer, which means we can use the utilization and the cost that will, by the way, continue to come down over time,” Hood said. “We’ve seen platform shifts before. And so, cost per tends to come down with scale, of course, and I think we’re starting with a pretty robust platform to be able to do that.”
Goldman Sachs analyst Kash Rangan asked executives about rivals, without mentioning the likes of Alphabet’s
Google and China’s Baidu
by name, that also have AI efforts under way. If everyone has AI, why would Microsoft benefit?
Hood insists that Microsoft will have some advantages. “To build the platform, we referenced it as the AI supercomputer, but that work has taken years and it’s taken a lot of investment to build the type of scale, the type of speed, the type of cost that we can bring at every layer of the stack. I think that actually is quite differentiated at the scale at which we operate,” she said.
“While there are lots of models being built and there’ll be lots of competition in the space, there aren’t that many people who’ve invested at the scale we have and at every layer,” she said.