The one thing we can be sure of is this: Tea--the Camellia sinensis plant--always contains caffeine. Even decaffeinated teas still contain some caffeine (though it's usually less than 3% of the original amount). In reality, caffeine testing is an extremely expensive, technology-intensive process, and most tea producers, distributors and especially independently-owned tea shops like Miro Tea couldn't dream of affording this costly sort of testing. In fact, the volume and range of scientific studies on caffeine content in different tea types, caffeine extraction over time and temperature, and the way different processing steps affect caffeine content is actually quite limited, meaning that there isn't really a significant body of evidence to make claims about these subjects ironclad in any way. The unfortunate reality is that many tea-related businesses deal with these ambiguities by taking shortcuts and liberties with the facts that are often blatantly inaccurate and misleading, not to mention the fact that they perpetuate misinformation in such a way that incorrect assertions like "white tea is low in caffeine" have become common "knowledge."
There are, however, a few general items that seem reasonable to say regarding caffeine content. Factors that seem to indicate high caffeine content in tea include: Higher quality leaves (generally the buds and first leaves), large amounts of nitrogen fertilizers, species of tea plant (assamica plants have shown higher caffeine content than sinensis), the condition of the leaves (broken leaves expose more surface area to the water), longer steeping time, and higher water temperature. However, it's not even as simple as tallying these variables--high quality leaves usually contain more theanine, a compound which has been shown to act in a way that is physiologically contrary to caffeine, for example. Another factor that I think is often overlooked is that many people report that the caffeine in different tea types (green, white, etc.) affect them differently. Combining these physiological factors with the dearth of scientific evidence on the numerous other factors, if we were to claim that, say, our Meleng Assam has more caffeine than our Sencha, it would be an almost laughable proposition (if it weren't so irresponsible).
So, what is the best way to approach this complex subject when customers ask questions? I honestly feel sympathetic for the tea companies who take the easy but inaccurate way out when spreading inaccurate information about caffeine--it's tough to provide ambiguous answers in a customer service situation--but honesty and transparency are key principles in good customer service relationships, so I can't condone oversimplification for the sake of brevity. I prefer to concisely explain that tea's caffeine content is much more complicated than popular sources let on, and that the number of contributing factors and the cost of testing make determination of caffeine content very difficult. If a customer is extremely concerned with avoiding caffeine, I recommend botanicals or rooibos--complete abstinence from caffeine is a sure thing. Otherwise, I'll recommend that the customer try whichever tea they're interested in the morning and take note of how effective the caffeine is for future reference. The one situation in which I feel confident describing a tea as "lower in caffeine" is when the tea is a blend with a significant amount of botanicals. Since we measure our tea by weight for each serving, teas like Genmaicha and Masala Mint Green contain, gram-by-gram, less tea leaves than pure tea blends do, so there will obviously be a lower caffeine content. These are about the most concise answers I can think of that sacrifice the least accuracy. Luckily, we also have this blog as another medium for a slightly more suitably detailed (though still not really scientific) description. If your scientific appetite has been piqued at all, I encourage you to check out the venerable Cha Dao Blog's past posting Caffeine and Tea: Myth and Reality, to which I've made general reference a couple of times in this posting. The article is pretty academically rigorous and it includes citations of the existing relevant studies and figures from their findings. Also of note is the ensuing discussion in the entry's comment section, which continues the myth-busting spirit of the article in several meaningful ways.
I hope this posting has been at least a little illuminating about this difficult subject--hopefully the next few years will see the publication of some more relevant scientific studies so the "mystery" aspect will diminish a bit.