8 Comments
Jan 29Liked by Sangeet Paul Choudary

> If users expect Google Bard to proclaim the one right answer, Google has demonstrated that it is sufficiently incentivised to present its own high margin solution as the best one.

I’m confused. If I’m using Bard to get one right so sorry answer from a Google search, how are they making any money from their ads? What is the high margin solution tied to Google Bard?

Expand full comment
author

I'm going to take Amazon BuyBox competition as an analogy here. One of the factors that helps merchants win the Amazon BuyBox is high transaction velocity. However, the more popular the product (i.e. the more merchant substitutes available), the more likely it is that the merchant will have to pay very high advertising fees to stay on top of search results on Amazon and hence drive transaction velocity. There is a demonstrated link between BuyBox placement and ad spend.

I expect Google is similarly placed and incentivised to tie the default answer on Bard with publishers who invest more in ad spend. Amazon doesn't explicitly tie the two (only doing it loosely through transaction velocity) but in a competitive market, you need to outbid competitors for visibility and Google already provides the mechanisms for that (e.g. allowing Monday.com to bid on the keyword 'asana').

Expand full comment

That seems self-defeating. If Bard provides “the best answer money can buy”, then it is no longer the One Right Answer.

Expand full comment
author

Google has already moved away from providing the best answers to providing a mix of (1) the best answers, (2) Google's own properties as answers, and (3) the most monetizable answers.

On monetizability, ad costs are based on ad quality (a measure of relevance) so there is some counter-balancing in the sense that the lower the ad quality, the higher the publisher ad budget will need to be.

Expand full comment

So... your argument is predicated on users *believing* Bard has the One Right Answer, even though they’ve totally sold out? That doesn’t seem sustainable...

Expand full comment
author
Feb 1·edited Feb 1Author

Really? It's worked for Google now for more than 16 years. Users continue to overwhelmingly use Google and 'believe' it is the better search engine compared to Bing. Arguably, there is sufficient evidence to show that Google has sold out a lot more and that the quality of search results on Bing isn't poor enough to justify the market share.

As Amazon demonstrates, tying can enable a seemingly relevant provider with a higher ad budget to rank higher than the most relevant provider without an ad budget.

Expand full comment

Yeah, for search where people can (and do) choose between the ads and the “real” results. Is Google going to label Bard as a “sponsored” result?

Expand full comment

Maybe Perplexity (and its deep-pocketed investors) can take down Google. I can see how its technology can be used in various, small LLM models to solve a vertical need.

https://www.nytimes.com/2024/02/01/technology/perplexity-search-ai-google.html

Expand full comment