February 07, 2009

E-Mail Strategy

Which of the following three strategies would you recommend to management?

Strategy #1:
  • Two e-mail marketing campaigns per week, untargeted, delivered to the entire list, no promotions or offers (THIS IS THE CURRENT STRATEGY).
  • List Size = 100,000.
  • Average Sales Per E-Mail = $0.10.
  • Average Opt-Out Rate = 0.25%.
  • Average New Subscribe Rate = 0.35%.
  • Average Weekly Demand = $20,000.
  • Average Weekly Profit = $7,000.
Strategy #2:
  • Two e-mail marketing campaigns per week, untargeted, delivered to the entire list, each with a free shipping offer.
  • List Size = 100,000.
  • Average Sales Per E-Mail = $0.20.
  • Average Opt-Out Rate = 0.20%.
  • Average New Subscribe Rate = 0.40%.
  • Average Weekly Demand = $40,000.
  • Average Weekly Profit = $9,000.
Strategy #3:
  • One e-mail marketing campaigns per week, untargeted, delivered to the entire list, no promotions or offers.
  • List Size = 100,000.
  • Average Sales Per E-Mail = $0.15.
  • Average Opt-Out Rate = 0.15%.
  • Average New Subscribe Rate = 0.30%.
  • Average Weekly Demand = $15,000.
  • Average Weekly Profit = $5,250.
Strategy #4:
  • One e-mail marketing campaigns per week, untargeted, delivered to the entire list, each with a free shipping offer.
  • List Size = 100,000.
  • Average Sales Per E-Mail = $0.25.
  • Average Opt-Out Rate = 0.15%.
  • Average New Subscribe Rate = 0.35%.
  • Average Weekly Demand = $25,000.
  • Average Weekly Profit = $5,625.
Strategy #5:
  • One e-mail marketing campaign per week, ten versions highly targeted only to the "active" e-mail list of 20,000 addresses, with free shipping offers given only to the 10,000 who like free shipping offers.
  • List Size = 20,000.
  • Average Sales Per E-Mail = $0.50.
  • Average Opt-Out Rate = 0.10%.
  • Average New Subscribe Rate = 0.20%.
  • Average Weekly Demand = $10,000.
  • Average Weekly Profit = $2,500.
Which of the five strategies would you employ? If you chose strategy #5, describe why you would choose this strategy.

11 comments:

  1. Kevin,
    I am a bit confused by your post. Option 5 seems to have components you favor(targeted free shipping, targeted email) but both profit and subscriber growth are worst of the options, and cost of creating 10 separate campaigns has to be part of that. Oh, and someone at C-level would say "do more than 20k!"

    Option two appears best in terms of net increase of subscriptions and profit, even though just doubling effort of Option 4 with better profit. That old soviet maxim that Quantity takes on quality all its own applies? But I thought you disliked blanket free shipping?

    Options 1 and 3 seem to stress a semblance of organic growth by not offering anything--just getting name out there but confused to why numbers for 3(single campaign) are so much better than option 1(two campaigns), when you in Options 2/4 you have opposite effect?
    I look forward to your explanation of this puzzler.
    Also, are numbers extrapolations from real data? These results definitely deserve more commentary...

    ReplyDelete
  2. I agree with the previous commenter that there are some mysteries yet to be revealed. On the face, I would choose #2, because it is tied for the best net list growth (opt-in less opt-out -- .20%) and the highest weekly profit.

    ReplyDelete
  3. Gentlemen, the results reflect what many retailers see when they do tests of this nature. And the results do not always match up with the best practices folks recommend retailers adhere to.

    ReplyDelete
  4. Anonymous5:15 AM

    While we haven't tested all 5 options, I can tell you that we have tested reduced frequency and reduced promotions and Option 2 always wins - which is why you see the explosion of free shipping, promotional emails with 6X/week frequency in your inbox.

    It is a bit of a prisoner's dilemma in that even though you know the "right" thing to do is get targeted and reduce frequency, you also know that (at least) in the short term, you can maximize profit by blasting away. You also know that your competitor is likely to make the same decision, causing you to lose market share if you go with option #5.

    I'd love to see case studies of companies that choose Option 5 and have found that it actually increases long term profit.

    ReplyDelete
  5. Hi Anonymous, it seems like a lot of people observe that #2 works best. #5 should work best, but seldom does, and there aren't a lot of case studies that prove mailing less often in a targeted way is best. Maybe one of our loyal readers from one of the e-mail vendors could supply three or four case studies.

    ReplyDelete
  6. Option 2 is the clear winner in terms of list growth, revenue and profit max. However, a key piece of the puzzle is missing, and that is the effect of this blasting on response rates. All of the options that send untargeted emails twice weekly should suffer from significant offer fatigue and a higher opt out rate. Unless you are Victoria's Secret of course!

    ReplyDelete
  7. Mix it!
    #2 gets the biggest profit, so if you don't have time to experiement, stick to it.
    If you have some time to experiment, make a mix out of #2 and #5!
    #2 Send out one mail per week targeted to non active adresses, every week with free shipping to 30%.
    #3 Send out targeted mails to sub-audiences as described, but offer free shipping to the 'non-free-shipping' targets 4/5 times per year.

    The targeted is much more work and in that time you can't do other things, and the untargeted has the highest profit, so we cannot leave that out.

    ReplyDelete
  8. Well that's a practical solution!

    ReplyDelete
  9. Bob --- you mention that you should see fatigue ... per anonymous, can you share with the readers any case studies that illustrate your point?

    ReplyDelete
  10. OK, I'll bite, even though it's not my job ;-)

    Case studies are largely conducted to demonstrate how a change brought benefits and not to demonstrate how not changing made things worse.

    Nevertheless, I dug out some studies a while back to support the idea that segmentation and targeting makes sense. See here. Not sure if you count them as valid, but I offer them up anyway.

    Having said that, I'm not arguing that any one approach is better than the other. I can quite believe, for example, that some lists will not fatigue, while others will. We tend to fall into generalisations too easily.

    More accurately, it's likely that a section of the list will not fatigue (your big fans) and a section will: identifying those groups and matching frequency accordingly would be cool.

    Also, fatigue might occur over a long timescale, one far bigger than most tests are conducted for.

    Finally, the concern with fatigue (which for the record I think is a valid one worth at least testing for) is a bit like climate change.

    Just because you didn't have a problem so far, doesn't mean it can't happen in the future as people's attention gets ever more shorter and inbox competition grows. (Assuming you agree with those trends).

    So once you find the option that works best in a test, I'd always add the corollary "for now" and retest your assumptions later.

    And email is fairly unique in that if enough people get fed up of your emails, you can cross a magic complaint threshold and find all your emails blocked at major ISPs.

    So, for example, you can bump up frequency and get more and more profits, right up until you hit that threshold. Then it's not a gentle decline but a catastrophic fall in profits. (My hypothesis: I'm just a lowly observer).

    And like andreas.wpv said, there are mix and match options here.

    No axe to grind, so interested in any other opinions.

    ReplyDelete
  11. Seems reasonable, Mark, thank you!

    ReplyDelete

Note: Only a member of this blog may post a comment.

Retail

Yeah, that's a lousy picture. Too bad. Today is essay day. If you don't want to read something long, stop here. I spend a...