Aside from firing an employee, firing a product is the worst part of my job. And yet at this time of year, we face our annual cycle of terminating underperforming varieties. This month’s victims are, brace yourself, our beloved Pearfect White and Sublime Mate.

Our whole product development cycle isn’t that different from planting a tree:

1. Before we plant any ideas (let’s call those seeds), we make sure they’ll be able to thrive in our ecosystem. You can’t plant a tea bush in Maryland (I've tried) and we won’t explore a beverage that can’t be certified organic. We also have to make sure that the drinks can taste great without a lot of additional sweetening. And finally, if it’s a tea, we need to make sure there is a Fair Trade certified supplier.

2.  If a concept starts to take root, we make up some samples in our global research and design facility, which in reality is our office kitchen (when it’s not also serving as our lunchroom). We begin exploring  names and artwork to help convey what we’re selling, and we start sharing conceptssamples to taste, artwork and packaging to respond towith our key retailers, brokers and distributors, who inevitably have good insights about how to tweak the name, taste or label image and language. This is also a smart way to make sure our sales partners have a sense of ownership over our sapling, and are willing to help nurture it.

3. Next, the greenhouse stagewe’ll expose our seedling to the lightwe’ll do a bunch of tastings of the product in and out of our office. And then we start to finalize our label language. Barry and I still write the back label messages, though we are glad to have more editors than when we started. We often find ourselves agonizing over the smallest detailsword choices (my straightforward language versus Barry’s sarcasm).

4) Finally, we launch the product. Our tree is now planted and has to face the elements on its own. We do our best to make a big deal out of the launchpress releases, introductory promotions and samplings, but ultimately consumers decide if our tree will thrive or wither.

Over the course of our product development process, which can run anywhere from three to nine months, we fall in love with the end result. We choose a UPC code (those digits on the back of the label with the bar code) that is the birthdate of one of our employees, and we send out birth announcements/sell sheets. When something doesn’t sell well enough, it represents a failure on our part. Either we didn’t design the product well enough(anyone remember Tangerine Green?) or market it well enough(Haarlem Honeybush anyone?).  It’s tempting to want to blame the retailers or the consumers for a failed product, but ultimately it’s our fault.

We inevitably receive impassioned calls and e-mails from frustrated consumers who can’t find their favorite drink. They ask why we can’t just keep selling the drink even though it’s not selling as well as our other drinks. In an ideal world, we’d never discontinue anything, but a slow-moving item takes away shelf-space from a fast-moving item. And since we don’t have unlimited shelf-space, just as in the plant and animal world, it really is a case of survival of the fittest.

We also see some beverages run their course. If a drink that was once a top-seller loses its mojo (we hardly knew ya, Kashmiri Chai!) or it gets cannibalized by one of our own drinks (Wherefore art thou, First Nation Peppermint? Yon Moroccan Mint has a lean and hungry look.), it’s very hard to revive. More than once we’ve tried to improve the offering and re-launch it (how about another Gold Rush Cinnamon label contest?) but so far our attempts to reinvigorate a withering plant have failed.

I take each discontinued product hard. And each time I hope that our customers will keep the next tea from the chopping block by buying lots of it.