Startups, please write better emails | Hacker News Reader

Startups, please write better emails

i.mhj.tc

102 pointsmikk0j38 comments

People don't write bad emails because they don't care. They really do care, but the emails that come from startups are often written with a mixture of fear of being perceived as spam, and a lack of understanding of how to connect with people.

Before I started an email startup, I would probably have written the same types of emails. It's amazing that once you start to focus on something, you notice the little things.

I taught a class a few weekends ago on how to write better emails (slides: https://speakerdeck.com/sudonim/write-emails-people-will-rea... ). I'm happy to help anyone (for free) who wants to do things better than they are now. Contact info is in my profile.

Yes, I think they certainly don't write bad emails because they don't care. They might write bad emails because they care too much. They may think about their product constantly, so stating the obvious about it doesn't even come to mind (and I generalize horribly here).

Good slides btw, worth checking out for anyone. Many startup email writers may be even too conscious of following the AIDA guideline and thinking that in order to grab the reader's attention, they have to open with the feature or the announcement, instead of reminding what it is that they do. Happy you address the 'noreply', too.

Great deck! Out of curiosity, one of your slides seems to say that plain text emails perform better than HTML (i.e. fancy) emails:

Plain text: $17,500, Fancy: $4,300.

A bunch of replies in this thread have people saying HTML emails perform better than plain text ones:

http://news.ycombinator.com/item?id=4856529

Am I reading any of this data incorrectly? Is this a case where the answer really is, "it depends"? Or is there truly one format that is better than the other?

Hey Mike,

Right now my position is that for the effort, you're better off writing good content than worrying about your template.

We've seen really strong performance from simple formatting (bold, italics) where the focus is on building a relationship with the user:

"Hi, I just saw you signed up and wanted to help you get started". This is meant to feel hand-written.

I saw that thread and rather than make unsubstantiated claims in response, I want to get a better understanding of when you should use a highly-styled layout and gather some data.

Many of our customers like to use our "light" layout ( https://raw.github.com/customerio/skins/master/Light/light_s...) for transactional emails and put their logo up top. There's a bit of styling, but it is really subtle.

Multi-column in my opinion is way out. It's a mess on mobile, and you probably shouldn't do it. A great example of a nice single-column, styled email is campaign monitor's newsletter.

http://campmon.createsend.com/t/ViewEmailArchive/y/78AABCA56...

Yes, the "fear of being perceived as spam" is certainly something we all have. I really feel like stating what our company does would be like "Hey I'm company X trying to sell you Y you don't care about".

Well, if you don't tell us why you are emailing and what your company does, then the emails just come off as:

"Hey, I'm company X and I'm not going to say much other than we've updated stuff based on feedback other people provided but I'll include you in that group."

You then go on to say:

"Version X.beta{4.3} is ready for release and you want to thank me for my support."

Finally....

"Follow us on twitter where we won't post anything because we don't want to spam you their either."

Now, I vaguely remember signing up to get info on something a long time ago, and you might be those people, but I don't have a clue. And now you don't even have the courtesy to let me know who the hell you are? How we are related?

No, no, it's even worse than that.

You are a company that proposed an idea, and I asked to be informed when more information was available. You've already got me listening. I'm interested. I want to be told why I should give a damn about you. Instead, you have the nerve to tell me that my interest is not worthy of your time and effort to spend a few minutes trying to sell to me?

"Hey I'm company X trying to sell you Y you don't care about"

I already asked to be sold! I already showed I cared! If anything, you've taken my first bit of feedback and told me to fuck off. I'm not worthy. Why?

Because you are scared.

And it gets better. You aren't selling to me sends off warning lights. So, why should I rely or buy from a company that has proven they:

1. Don't want to sell their product. 2. Don't listen or trust my input. 3. Don't believe in what they are doing to try and sell their product. 4. Don't even want to tell people who are interested about how to get their product.

I want to support cool new ideas. I've got money. I'd rather see it go to a cool new startup then a frappe-latte-mocha-chocolatte-ino.

So stand up and be an adult. Be a leader. Say "Here is what we built, here is how it helps you, and this is where you can buy it," because if you don't do that, you are really saying "I don't care.

---

Then I go to your profile here on HN.

http://news.ycombinator.com/user?id=louischatriot

"I finally decided it was time to launch my own company, with Charles and Stan."

Unless you haven't announced it yet, why aren't you talking about it? Why aren't you sharing it? Is what you are doing worthless? Then it doesn't matter. But is it something worthwhile? Then share it!

Sell me! I asked for it!

Those slides are fantastic. Seriously great content, especially the "start the e-mail with the template" slide. Thanks for linking that.

On the second "simple or fancy" slide, what were the results of the data on that? Same (general) results as the first slide?

Hey Charlie, the second slide was just another example showing contrast between plain and fancy -- more of a talking point during the class. Both of them were reader submitted so I didn't have any data.

It's something I'm interested to test a little more. In general, plain text is a pretty safe, high performing option. However, I've heard that a well designed HTML email has done better in A/B tests. I want to find the proof.

Awesome. Yeah, I feel like it's the kind of thing where people opening their data could be really useful for everyone. (Though I recognize the difficulty in accounting for all the variables in terms of audience, message, design, etc.) I signed up for your newsletter; if you ever do scare up more data on it, it could make for a great note.

I wish programmers would also follow this advice when creating their GitHub repositories and project sites because they can suffer the very same problem.

It's pretty often I hear about a library or project, hit its homepage, and I can't figure out what the heck it even is without digging around.

The magic phrase, as used in this article, is "[thing] is [describe thing]". Tell me the name of your thing and what it is. Once you've got that paragraph out of the way, you can go to town!

Completely agreed. Like pointed above by venus, the "best for" format may annoy some people and is not necessary in all cases. Your more generic example works as a better template.

> The magic phrase, as used in this article, is "[thing] is [describe thing]". Tell me the name of your thing and what it is.

And "[thing] is [describe thing]" does not mean "[thing] is [like other thing but for X]", because I have no idea what other thing is either.

Slightly off topic: the introduction the author recommends using, "(name) is the best (product category)", was slightly annoying a couple years ago when I first heard it and is now just a full-blown obnoxious cliché. Whenever I hear a startup type guy talking about how their product is "the best" blah, I automatically append "according to me". You may as well just say "my favourite".

Startups, please don't introduce your company as "(My company) is the best (what we do)" - that's for the market to decide, actually, and for you to just claim it up front is presumptuous, at best.

Thanks Venus. No I don't recommend that across the board. I did use "X is the best Y" 2 out of 5 times in the examples (both are directly from their value props on the site or on Twitter bio). Personally, I don't find it annoying anymore but I do agree that it is devalued. Do I believe they are the best X for Y? Of course not. Maybe I'm more lenient since think of it more in terms of vision and ambition.

I'm all for vision and ambition but there's a difference between aspiring to be the best and claiming to be it already! One is a laudable goal; the other comes across as wishful thinking.

Imagine if you opened a new restaurant and on day one your sign proclaimed it as offering the best food in the world. Or a tiny little brand new startup furniture shop, "X furniture is the best furniture for your home or office in the whole world". Can you imagine the utter ridicule such a claim would invite? Why is a software startup any different?

Claim to be the easiest to use, most elegant, most cost effective, fine. There's nothing wrong with selling your strong points. But just going for broke and all-out claiming to be "the best"? Maybe I'm just a cynic but I imagine some little kid putting on a superman costume and claiming they're the strongest man in the world. Sure you are kiddo!

I come from a marketing background, so I guess I'm both numb to and guilty of that kind of text. But I don't see how "most elegant" or other superlatives are different from "best for X" though. Both are equally difficult to prove, both may be completely subjective and contingent.

I like the kid in the superman costume analogy for startups and big ambitions. I'd cheer him on. If he believes it he'll try to act the part, so the more I'll be inclined to believe him, too (though I might stop him from trying to halt a truck).

> I don't see how "most elegant" or other superlatives are different from "best for X"

I've already argued about this way too much but let me give you an example : P

If I asked you what are your strengths, what distinguishes you as a person, what would you say?

Now if you're like a normal person, reasonably modest and yet confident of their merits, you might say you're an ambitious person who's serious about their goals but knows how to have fun, you're an excellent communicator and listener, you're a good friend and you bring a genuine passion to everything you do.

You'd come across as a really cool guy.

Now imagine if you just crossed your arms, assumed a smug grin, and said, "well, I'm the best".

What do you think people would think of you if you did that? Well I don't know about you but I'd grin and nod and think "this guy is a fuckin' douchebag".

I'm not trying to say I think every marketer who ever wrote "X is the best Y" is a douchebag; I'm saying that claiming to be "the best" is a very, very big claim. Coming from a giant of software it would be questionable; coming from some tiny unknown startup it is simply laughable.

Stick to pointing out your merits, strengths and focuses. Don't make a claim to be all-categories, all-uses, undisputed best thing in the world. You are almost certainly not, and to this engineer, it raises questions about what kind of person would even make such a ludicrous boast.

He reckons a sentence like "Learnist is the mobile product for curating and following complete learning experience" is worth adding to an email? Good grief! I actually have less idea of what it is than I did before I read that, and I've never even heard of it. If you're keeping track, that's negative knowledge. No, I didn't know that was possible either. You live and learn.

Seriously, though: if you have a business, and you can't explain what it's about without making people's brains bleed, perhaps you should go work for Google instead.

I'm keeping track. Thanks for the feedback - obviously not a good description for everyone. Would have worked for me, though.

Maybe I should've stayed at Google then? :)

My suggestion (which is worth exactly what you paid for it) is this: find someone who has no contact with the IT world beyond, maybe, a passing familiarity with mobile phone apps and Google Mail. Tell them you want a one-sentence description of your own business. Give them as much information as they think they need, explaining everything in sufficient detail so they feel they understand it. Then ask them for their one sentence. Write it down. Do this a bunch of times. Finally, take all the collected sentences to someone who does understand IT but knows nothing of your business, and get them to condense them all into a final description.

I strongly suspect the result will be an improvement!

That's a great suggestion. It will certainly be an improvement. Just explaining it to a wide variety of people can help your thinking about it. If you get outsiders to describe it for you, even better.

It isn't just emails in particular. Think for a moment about some fancy sites you've come across - say a startup advertising their service/product. How many times after being fed a plate full of bells and whistles, you still don't know what they actually do? It tells you how wonderful their product is without telling you exactly what is that product.

But this does make sense, because if they put things in plain words most of them will be "This is another social platform", or "this is another cloud service", "another image processing tool", "another education software". That doesn't sound very attractive, and they're well aware of it. The trick is to first convince that this is something wonderful, then subconsciously people will perceive it as something different from all existing services - and so willing to try it out.

However, most users are getting increasingly sophisticated and skeptical, and this sensible strategy sometimes backfires.

Everyone, please write better emails. Provide me with a single, clear call-to-action (aka next step) so that I "level up" and become more awesome. Side effect: I'll use your product and learn about your company.

Remember Kathy Sierra's advice: Don't buy this because we kick ass, buy this because we want YOU to kick ass.

http://headrush.typepad.com/photos/uncategorized/2007/04/06/...

I saw Kathy Sierra speak at BoS 2012. After her talk, we wrote on our wall in the office as a constant reminder our mission is to: "Help our customers be badass at emailing their users".

Copyrighting isn't easy, much the same as writing isn't easy. There's an inherently different mindset than most technical people are used to: putting yourself in your target audiences shoes and trying to fulfill their wants/needs/questions.

Writing better emails comes with practice and some good ol' analytics checking on user engagement. Also, the title is as important as a blog title. Study up on words that tend to drive engagement and curiosity.

Well put. There are many services, like Mailchimp, that allow you to send a portion of your messages before sending out all your invites (basically an A/B test before unleashing ALL your email with one untested subject line). I don't think startups always think through their open rates to the extent they should.

The body copy is a whole different story, but you nailed that in your post.

A/B testing is a great thing to do with email. On the point of Mailchimp, they also allow you edit the little preview text box - this is what shows as the preview snippet in many email programs and is a good place to remind the recipient what's going on (very briefly, though).

Cool, thanks. I actually don't use the service as of now, but I've been over some of the features. I will probably use that service or a similar one in the near future.

This that annoys me the most - it seems a few startups don't know to include unsubscribe links in their emails any more. Two examples from just this morning:

http://cl.ly/image/0R101l3O101Y

http://cl.ly/image/34020b0m3n1m

Now yes, with ShiftEdit they tell me I can delete my account, and probably in both cases I could email them asking to unsubscribe me - but it's an extra effort I shouldn't need.

I have another service (to be honest I've blocked the name from my brain at the moment) - it was some spammy twitter app that continues to send me emails via a personal account rather than the company - with no way to switch them off - I've emailed them a couple of times asking them to stop, but they don't - so I just mark them as junk now.

One thing which worked very well for me is focusing only on one thing per email. Startups always have this conflict of trying not to spam their users too much but still making sure they know everything. The outcome is that every time they do send emails they try to include e-v-e-r-y-t-h-i-n-g in it. I struggled with it a lot as well, at one point we decided to dedicate each email to one new small feature, asking for a certain feedback, etc. Even when we released a new version we focused only on one thing. I think users find it much easier to read and digest. It also increases the chances your users will understand what you want them to do :)

I've definitely experienced this on the receiving end, both with startups and new products, but also with random mailing lists that only send out updates sporadically.

A good rule of thumb for ANY mass email communication is to lead with a short explanation of why they're receiving the email: "Thanks for signing up for updates from Newproduct, a new platform for knitters to connect online."

Remind them 1) they signed up, 2) what they signed up for (updates), and 3) who you are (descriptively).

Incredible. This is exactly what I thought when that Cue email popped into my inbox this morning. "What the hell is this?"

It was that email that prompted the blog post. For the Cue team's credit, they got in touch afterwards and we exchanged a couple of emails about it.

PeoplePerHour I'm looking at you! Your emails are awful, border on offensive, and I can't reply to them. Finally unsubscribed, which means I'll probably forget about their service /even though I actually like it/.

Great insights. Need more of that!

Thanks!

I thought the exact same thing the other day. I have deleted many startup emails recently, only because I didn't know wtf they were all about, and didn't want to sign up just to find out. Example: "Startup X. It was a long way bla bla but.. We have finally released!", like I had been waiting for them all this time.

That's the key thing: "like I had been waiting for them". No, we haven't been waiting for you, sorry. So please remind me what it was you do again?

Reply on news.ycombinator.com