It's likely that the economics of this don't work out. Most people actually do their online flower ordering by going to Google, searching [flowers], and clicking on one of the top few ads. (Indeed, I know about this because [flowers] was a very common test query we used when working on Google Search - it shows so many ads that if your new feature is going to break layout or look awkward on any query, it's probably [flowers].) In order to get people to click on the ad, you need to do your own branding.
Flower ordering has the same problem as weddings, funerals, job searches, ordering a plumber, etc: it's an event that you do maybe once a year, max. When you need the product that infrequently, it slips out of the consumers' mind, and they don't remember brands. So you can only build a viable business when the product costs enough that you can spend large sums on brand advertising (eg. cars, clothing, divorce attorneys, diamond rings) or large sums on search advertising (eg. [flowers], [mesothelioma], [digital cameras], [bingo cards]). Indeed, one of the main reasons Google has become huge is that they function as an aggregator for all the "infrequently needed, but too cheap to buy superbowl ads" products.
You're welcome to give it a go, though. The local florists would love you; many aren't terribly thrilled about Teleflora stealing all their customers and treating them as a commodity backend provider. It's just that you'll likely find customer acquisition costs way more than you bargain for.