Fair enough - but it is not mandatory for a Monero user to subscribe to the subreddit. All they need to do is have the client, so in many ways that is the most important metric to get an idea about adoption.
Quoting the Monero Missive - from July 15th to August 3rd there were 15,000 downloads of the Monero client.
It is true that not everyone will subscribe, but if you assume that the percentage of new users subscribing is similar between the two populations (the batch of users starting using Monero a month ago, and those starting now), which does not seem like a bad assumption at first glance, then the percentage of growth in subreddit subscribers should not be too dissimilar to that of client downloaders.
Another possible confounding factor would be whether a new client was released/advertised. Shortly after that point, there will be a glut of downloads, not only due to the increased awareness, but from previous users of Monero merely updating to the newest client, artificially inflating numbers (not that I've no idea if this applies in the time frame that was looked at).
Estimating growth using various methods should allow you to get a "middle ground", or at least a better estimation of whether your estimate is likely to be accurate (all calculated growths cluster around a similar value) or not (wildly disparate estimations).