This very large and very carefully cleaned dataset shows that the orders are clearly related to the Bulk Message having been sent; if nothing else, the dramatic falloff over time illustrates that case.

Not all of these messages were meant to drive non-club sales; for example, surely hundreds of the 4,894 Bulk Messages included in this analysis were meant instead to remind members about customizing their club, to pick up purchases already made, to resolve declined payments, etc. Although these messages are included in the total number of 4,338,837 recipients, it’s important to note that the club-related sales they generated are not included in the attributable orders above. Even given this extremely conservative analysis, the average Bulk Message still generated $11.67 of non-club sales per recipient!

Incredible!

Other Learnings: When’s the best time to send a Bulk Message?

With over four million Bulk Messages sent in 2023, we have a broad statistical basis to analyze send time and its effect on subsequent orders placed. First, a word of caution: the best time to send a Bulk Message is when it makes the most sense to do so. This data is only useful when you have several times that might suit the content of your message and want to pick the best among them.

With that being said, weekdays performed better than weekends and Fridays performed slightly better than the other days of the week. Like opt-out rates, order value by send time was weakly correlated, meaning it did vary quite a bit from winery to winery. That further reinforces that the most important consideration is the timing that makes the most sense given the exact content of your specific message.

Beyond just which day of the week, you may be wondering what time of day is best to send your Bulk Messages. Statistically, the “best” time is 3:07 PM Pacific Time.

Before you get out your stopwatch for precision timing, remember that it’s not possible to send all your text messages at once! Businesses are rate-limited by the telephone carriers and most wineries are allowed to send about one text per second. There’s also no way to guarantee someone receives your text message right when you send it; if that consumer has their phone in airplane mode, for example, they’ll receive it later when they reconnect. 

This graph shows order attribution per recipient by time of day the message was sent, bracketed by hour. 

For a more realistic view, imagine the graph with all those points rounded out and connected. 

It does clearly show it’s best not to send your Bulk Messages too early, or too late, and that targeting the mid-afternoon is a safe choice.

Other Learnings: Should you include an image, or not?

With this much data, we were able to find that the Bulk Messages that did not include an image resulted in a higher order value per recipient than those that did. Does that mean you should never include an image? Nope. It’s much more nuanced than that.

The correlation coefficient was low, meaning there was quite a bit of variation from winery to winery. That makes sense when you consider that not all winery’s customers are the same and that the specific content of some messages might lend themselves to imagery more than others.

What you can learn from this data is that you probably shouldn’t include an image in every Bulk Message. Review the content of your message critically. If an image makes a big difference, then consider including one. If it doesn’t seem to make a striking improvement, leave it off. Images also take up a lot of space in your message, automatically make the message an MMS (which can be more expensive to send than an SMS), and may encounter deliverability issues that a simpler, text-only, message might avoid.