The impetus for this post is a recent Return Path (RP) article on the clients’ e-mail marketing habits. There, they found that a majority of marketers continued to “sent[d] email at a steady, high frequency for a 19-month period, despite a total lack of response from the subscriber (no opens, no clicks, no purchases).” This was no surprise to me as it can be quite difficult to manage complicated frequency / timing / creative optimization e-mail programs. RP continued to go into all the woes and drawbacks of such a practice, but an seasoned e-mail marketer should already know them. The name a few, mailing inactive consumers is BAD because:
Read more…
There has been a lot of buzz recently around net neutrality and what it means for businesses and consumers. Net neutrality effectively means that content providers and ISPs should work together blindly. For example, an ISP like Verizon should not choose to limit bandwidth to or filter Google websites because it wants to charge Google a fee or because it disagrees with a posting slamming Verizon’s pricing on Google Buzz. Vice-versa, Google should not restrict Verizon consumers from visiting its websites or de-prioritize their requests because Google disagrees with a policy of Verizon.
Read more…
Digital marketers frequently discuss tactics for increasing the social engagement of their target audience. Much work has been done to show that Facebook fans and Twitter followers add more than just goodwill to a brand. For example, check out this earlier post. However, many companies have struggled to develop accurate valuations for the consumers that engage socially. Accurately measuring that value is of utmost important to companies like LiveIntent.com and to marketers that need to justify ad spend through ROI.
In a recent post on Marketing Vox, the author attempts to highlight a strategy that could be used to determine fan/follower value. The step- by-step approach to the calculation is as follows (directly sourced from the article):
- Run a promotion and offer a discount for a product.
- Count how many people redeemed that discount.
- Calculate how much profit made from that redemption.
- Determine how many people returned to buy again after the promotion.
- Incorporate core data you already know – in this case, how much the average person will spend over his/her lifetime with it.
- Combine the two to determine how much that newly acquired fan will spend with the company over his/her lifetime.
- Voila – you now have an accurate idea of that customer’s value to you as a business.
However, there are a couple of major flaws with this approach. First, the author fails to recognize that many of these people may have purchased a product without the coupon naturally. They are, in fact, socially engaged with the company. Second, the author fails to differentiate between consumers that join independently v. those that join for coupons. I suspect that one might find that the former are more valuable than the latter. Segmenting on that data point could be a great way to differentiate angel v. demon consumers. Third, by incenting consumers to purchase this product you probably cannibalized the sales of another product; consumers have a relatively fixed demand (with regards to quantity) for TVs, etc. In any case, while the high-level strategy is reasonable, execution and analysis are much more difficult than the article leads one to believe.
I have spoken a bit about privacy over the past month and how we are likely to see regulation within the industry. A fellow blogger, Eric Goldman, just published an update about an AOL case related to the public distribution of AOL search data in 2006 that contained sufficient information to be considered personally identifiable. Although the case is just beginning, the court opted to dismiss AOL’s request for judgment on the pleadings, making way for a continued legal battle.
Read more…
Facebook has posted information pertaining to its SPAM filtering system that is applied to internal messaging. The full blog post can be found here. Most of the methods they use to identify SPAM messages are also found in e-mail filtering. Here are a few examples:
Read more…