Hacker News new | past | comments | ask | show | jobs | submit login
How To Lose 1% Of Your Customers With Just 1 Email (onstartups.com)
78 points by yoseph on March 28, 2011 | hide | past | favorite | 43 comments



This is a classic case of paying attention to the wrong metrics. The unsubscription rate went up. But so did the sales. And yet, the author is paying attention to the unsubscription rate. Not the sales.

(Yes, it would make sense to pay attention to the unsubscription rate if the author shares the historical data of those unsubscribers. Did they make a lot of purchases in the past and the long salesletter is just bugging them off - which is reducing the "life time value" per subscriber?

But I don't think that data has been collected. (Or the author would have shared it with us in my opinion...) So - this to me feels like a decision made keeping the wrong metric in mind.


Point well made.

I do agree with you, especially when considering the life cycle of the user. Maybe NK never looked or calculated the average member value of his users. Some of these users might have reached the peak of their member cycle and decided to unsubscribe. It could have also been (as you mentioned) that the long text had "turned them off".

Nevertheless, the important factors in which he should focus on are the following: - CRM - keeping a good relationship with his user base - Sales - number of joins/sales made by the campaign --> which would transfer towards average revenue per email - Maintaining a good quality list of users who are truly interested in his product --> it's all about quality over quantity, and not the other way around.

Lastly, the most important factor which NK did pick up, was that there was a surge in unsubscribes. Why was there a higher than average number of unsubscribes? He looked into it, and attempted to "win back" some of those users by sending them a personal email. --> I personally would have added a question asking them as to why they had unsubscribed (to find out the truth behind the sudden spike in unsubscribes).


He never really explains why the first 2 long-form emails didn't result in such a backlash, but the third one did.

Was it, indeed, because the deal price was a lot lower than the first two, as implied by his later "solution" of using long-form emails for larger priced deals?

All in all, it seems that this piece was written for an audience that already knows what's going on, rather than for us. That certainly isn't a knock or a problem, though.


I unsub'd after that specific email with a comment that it was too spammy. He got back right away saying:

"Totally agree.

Working on toning that down big-time.

Hope you'll give us another chance some day."

Me: "I'll give you another chance right now, keep it Hacker News grade, and I'll stay. :)"

Him: "Agreed. Short and sweet."

So I resigned up. Lesson learned: You can repair damage. Some customers know you are human and will give you another go.


If I were the one sending out these mails I'd have been surprised that some people equate message length with spam. I'm not saying it is wrong to do so, I've just simply never personally associated email length with whether or not something is spam -- for me the problem of spam arises when single sources begin to email me too often resulting in an inbox with dozens/hundreds of messages to sort through a day. The actual length of each of those emails is basically irrelevant to me, I either care about them or I don't and I know that after reading the first paragraph.


"Imagine... If we sent 100 emails then we’d be out of business."

Not quite. Then you'd still have 36.6% of your initial customers.


Does he really needed to finish the article with “Sign up to AppSumo.com to see what our new emails look like.”?

It feels so cheap. He should have added the e-mail he was talking about, I use AppSumo and I remembered the e-mail in question, but I bet many people didn't (and they will not register because of that cheap plug!).

edit: punctuation and grammar.


I think that a lot of entrepreneurs leave money on the table because they're scared to do something that they think feels cheap. In a case like this, it's probably a tiny minority that would be turned off by this. I'd be absolutely shocked if having a blurb like that at the end of the post would have a negative effect.


It feels cheap not because the blurb but the way how it's written. What I would have done is providing a copy of the e-mails he's discussing and at the end of the article add the plug. "Join AppSumo.com if you want to receive our infamous e-mails stuffed with great deals". (Obviously I'm not a copy-writer)


It would be good to see the email that caused the ruckus... or did I just miss it in the post?


http://images1.bingocardcreator.com/blog-images/hn/appsumo-m...

Posted as an image to preserve the distinctive formatting. Continues in that kind of vein for a while.


Thanks! That definitely adds context to the post. It's funny but I can definitely see some people not caring for the type of humour.


Well, if that's the magic email number 3, then I think maybe he lost all the subscriber who had been stalked or had been arrested for stalking. In either case, I can guess those folks didn't find it very funny... The rest of us might.


haha, definitely looks like spam


I dug around my email box and I think this is the one: http://www.errant.me.uk/code/snip/10

It was long. And very odd.

I'm kinda used to that from Appsumo by now so it didn't really register for me. Truth be told, I just clicked through to see the offer straight away and ignored the rest as TL;DR. Reading it now.. the nicest description I can come up with is "rambling"

EDIT: Patio11 has a point about the formatting :) so: http://errant.me.uk/appsumo.html


Yikes. I suspect I'd be one of the people to consider unsubscribing after receiving an email like that.


Wow. Kind of reminds me of a cross between http://www.youtube.com/watch?v=4YBxeDN4tbk and http://www.timecube.com/


Looks like the more recent ones have a TL;DR section right upfront.

http://pastie.org/1726812


Wow. In what universe is this considered good copy writing?


You should set your page's charset to UTF-8.


I don't think you missed it. 848 words in the article and yet no mention of the actual email that was such a problem.


If you change your format, you can expect some temporary attrition, but you can't extrapolate that to a constant decline until, well, it becomes one.

I'm pretty potty mouthed on Twitter and have lost quite a few followers over the years but I look at it as guaranteeing that the people who do stay following are exactly the sort of people who I want following me. In the same way, it seems like the sort of people who buy things from AppSumo's offers are the ones staying on the list.. so getting rid of the whiners and non-buyers is a good thing.


I didn't learn anything after reading all of this, somewhat disappointing.


This is what the email looked like: http://www.errant.me.uk/code/snip/10

He also gave tips to get subscribers back.


A few observations:

I dread each time my inbox lights up but recognized that 60%+ of our business is driven through our email list. Our AppSumo.com emails for the most part were always short and sweet, just the way I liked them.

Then why the change?

I wanted to hire a real copywriter so I can focus on other things in the business. My friend Neville is a funny writer and had been extensively studying / practicing copywriting best practices.

Doesn't being a writer imply having studied and practiced writing? And what are "copywriting best practices"? Something tells me Neville is a factitious fall guy.

Usually we see < 0.5% unsubscribe rate

A 2x increase isn't that big of a deal and doesn't warrant the urgency eluded to in the title.

* The funniest thing about it?

This email was our highest conversion to buy ever. Ah, so this post isn't about losing customers, it's about justifying and excusing the new long form sales emails.


"And what are 'copywriting best practices'? Something tells me Neville is a factitious fall guy."

They laughed when Neville sat down at the keyboard, but when he started to write...

In all seriousness though, read On The Art of Writing Copy by Herschell Gordon Lewis, and also Tested Advertising Methods by John Caples. By reputation they are the two best books on copy writing and advertising ever written.


Best practices these days tend to involve putting things in CAPITAL LETTERS, and occasionally bold or italic, and BEING A COOL DUDE, and getting real, and adding ten foot high 'Click Here To Purchase' links.

I'll check those books out to see what advice they give, though; personally, I learnt copywriting by.. well, writing copy.


Haha, I love that line. Thanks for the references.


I stopped caring about appsumo because the deals started to suck and be products using appsumo to make sales, whereas before it was products I wanted and the discount made me think I HAVE TO GET THIS. I don't read their emails any more :(


I unsubscribed when I received the email in question. Some problems with this particular email:

1. Stalking people does not resonate with me. Bad lead.

2. What are you selling? Tell me what before why. Seriously.

3. The attempts at humor were really bad.

But the reason I was ready to unsubscribe is that I'd been noticing for a while that the AppSumo deals were more like extended free trials than actual cheap stuff. The spammy email pushed me over the edge.

I still follow the RSS, and sometimes click through to see the email-signup-wall.

By the way, I did buy one of the AppSumo deals early on.


So the new email has much better conversion but alienates a minute percentage of the user base ? And you choose to go back ?

Always thought its better to have less great customers than more bad ones. Guess I am more sales oriented that email-list-size oriented.


They're a deal site right? So essentially their conversions are for the third parties, not for them.

It's possible that their sales to the third party companies depends more on their list size than on their conversion rate.


One thing I definitely have to say, is that NK really had some courage in trying to re-email the users who had unsubscribed. It is common best practice in email marketing to not contact unsubscribed users again, but kudos to him for testing out the concept of re-engaging unsubscribed users and specifying in his email that he made a mistake and would like to get a 2nd chance. Risky business, but I call it a success when he managed to win back some of the customers who had initially unsubscribed!


How long before we see the dark pattern of an automated script emailing unsubscribers with a "this is a personal message from Founder Name to apologise if you found our emails unhelpful. We had a few people unsubscribe recently and so we're working on improving our tone - feedback is welcome! If you do ever feel like coming back..."


So essentially what we have is a 1% unsubscribe rate for an e-mail, which is above his .5% average but it's still within the range of error and many things could effect the unsubs except the email content.

Maybe his e-mail list is growing and after getting a few e-mails people realize that this isn't what they expected. Or the expectations they had for the content weren't set properly. Saying that funny/longform e-mail is that what caused the sudden unsub spike is jumping to conclusions.


I wonder how many of the unsubscribers from message number three had just had enough of it... Having forgiven the first two... a cumulative effect.

I know that I have been very happy on several mailing lists of entites wanting to sell me stuff, but when the messages get too frequent, too long, too spammy or too far from my stated interests... Unsubscribe. I forgive a few indiscressions, but when I see this abuse is a pattern, I unsubscribe.

We use paper and toner, etc. We can buy extra on a good deal, so I was on Staples list. Started out a message every few weeks, then weekly, then daily... Unsubscribe.

I was on NewEgg's list. We use cases of disks and RAM and other commodities, so we can take advantage of exceptional pricing if the chance presents itself. We also sometimes must go to alternate sources if our primary suppliers are out of stock. I am on their list, and even told them what I wanted to see. However, they still tell me about microwave ovens and watches... Much less frequently now that I have told them my preferences, but they are still in danger of losing my subscription. They are on the line... But I'd have canceled already if they sent me offers on microwave ovens more than once in a very great while.


I found this interesting timing, as I just unsubscribed from AppSumo today. Today's email was a huge dive into web-copywriting-drive-conversions-awfulness with all the stuff I really loathe about that corner of the Internet. Even when the product in question is 'free'...

I understand the value in trying new stuff, but it's the old early-adopter problem: if you have a certain loyal userbase, do you risk doing stuff that they are too savvy to pick up on (and might backlash over), or do you pander to the huge new userbase you got from a recent promotion? Eventually you have to leave the fluffy world of appealing to HN readers, but there are ways and ways.


> On the morning of March 2nd we sent out an email to 40,000 people and subsequently lost 1% of those people over the next 24 hours. > Imagine... If we sent 100 emails then we’d be out of business.

Without knowing the context of the e-mail, or why that 1% quit, this is basically the equivalent of news hosts saying, COULD YOUR CHILDREN DROP DEAD TOMORROW? Tune in at 11 to find out.


I've learned (well, maybe) the hard way that if you say stupid things, it damages relationships. Now the reason one might keep doing so is that usually it's a lot more fun to say questionable things...

Alas. But yes, you have to be very careful with what you say. It's many times harder to overcome a negative impression than to create a positive one.


I certainly noticed the changes in the emails and honestly thought that they are A/B testing with different formats. I actually remember giving a thought of if this format (really long and nonsensical) really works, and whether they cut back a bit in the next email. That 2nd March email was quite exciting read nevertheless.


What about short explanation of the deal at the top, and a long funny story after that?

I too found the long emails easy to dismiss. I don't even know where to look to find out if I need this real quick - and I'm not going to invest the time to read it to only find out its not of any interest to me.


Why not have two mailing lists, one tweetable in content and the other funny? Start people off on the tweetable one, and give them a link if they'd like more info. And give them a link back, if they choose to do so.


I wonder what % of customers will be gained from this article?




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: