But when do you respond to it, and when do you ignore it?

1. The Pure Vitriol-Filled Name-Calling Email

Example: (I don’t think I need to provide an example.  We’ve all seen these.)

Verdict: Ignore.

No good can come from responding to these.  Best case scenario, you respond rationally and embarrass the hell out of the person who now regrets sending such an unprofessional email.   That’s the best-case scenario.

I’ve seen lots of worst-case scenarios where this devolves into an otherwise reasonable product manager sinking to the level of the original sender and then the whole exchange gets posted on someone’s blog and everyone is embarrassed.

2. The 90% Vitriol-Filled Name-Calling, 10% Actual Valid Issue Email

Example: “I can’t believe you f***ers lost my data after I spent 15 minutes typing it in!!! This is the worst piece of …”

Verdict: Respond.

I actually think it’s a great idea to respond to these, provided that: a) you delete all references to the rude parts and b) you reply to the actual valid issue with neutral, objective language.  (“You’re right, it was unacceptable that you were unable to find your data.  It’s actually not lost, though, your saved draft can be found [here].”)

By deleting all references to the rude parts, you allow them to pretend that you didn’t see it and they didn’t write it.  But if you can’t be neutral/objective, just delete it.

3. The “Whatever, You Suck” Email

Example: “your pricing sucks I’m outta here”

Verdict: Ignore.

There’s a complaint here, but no substance to respond to.  What can you say — ‘no it doesn’t'?  No.   You have to just ignore these.

Note that if there’s a reason or some additional supporting details “Here’s why your pricing doesn’t work for me…”, then you can and should respond.  Even if all you can say is “I’m sorry we can’t do more but we appreciate your feedback”.

4. The Frustration-Laden Tirade Email

Example: “First I tried to search for a message I’d sent but kept getting no results, so I had to page through three hundred messages individually, then when I finally found the right one I tried to print and the page stopped responding and logged me out and then…”

Verdict: Respond. (preferably via phone)

It is much easier to call these people, sound sympathetic, and let them vent for a few minutes without interrupting.  Then, figure out what they wanted to do and walk them through doing it.

If you don’t have a phone number, try to do the same thing via email – agree that it was frustrating, agree that that’s unacceptable, and then figure out what they wanted to do and walk them through doing it (or, if possible, do it for them.)

These are the people who, if you ignore them, will hate you and spread negative feedback about you whenever they get the chance; but if you respond nicely and help them, can be turned into your best WOM allies.

5. The I’m Not Using Your Product Again Until You …  Email

Example: “I keep waiting for you to support [feature X] but you still haven’t.  I’m going to have to stop using your product and look for another solution because [feature X] is totally critical to my business.”

Verdict: Respond (unless it’s a repeat offender).

It is tempting to try and convince the person that they should continue using your product.  It is also ineffective (most of these people will continue to use your product, albeit a bit angrily).  If there is a workaround to accomplish the same goal that [feature X] would achieve, tell them about it.

If you will never, ever, ever build [feature X] because it makes zero sense to your business, tell them that as well!  (Most people are reluctant to do this, but I’ve had almost exclusively good conversations that started from being really honest and saying something like: “We’re not planning on building X, and here’s why.”

The one exception: if that same guy keeps sending you the same email about [feature X], you can ignore them after email #2.

6. The Email/Tweet/Blog Post/Comment That Makes You So Mad That You Really Want to Write Something Rude Back

Example: will vary by person.

Verdict: IGNORE IGNORE IGNORE (and then think about it).

It doesn’t matter if there are totally valid points that should be addressed, or the person is completely factually wrong, or even if you feel like this is borderline slander – if you are angry, do not respond.

If you have done your job building relationships with your other customers, the odds are high that someone else will respond on your behalf and correct any misinformation.   Or you can ask a more stoic coworker to come up with a neutral response.

But even if not, think about the options:

1. Angry ranter writes comment, no one responds (clearly no one agrees with them, so you end up looking fine)
2. Angry ranter writes comment, lots of people come to your defense (you end up looking even better than before)
3. Angry ranter writes comment, lots of people agree (clearly you have a problem, and you should focus on fixing it versus debating it)

Safer to ignore (and see #1: avoid being sucked into an argument that can only make you look bad.)

Of course, for all of us, our goal is to get as little of this email as possible.  But until then, keep growing that skin thicker, and feel free to hit the ‘delete’ key sometimes.

If you feel the need to say this, then you have screwed up.

Maybe your copy was confusing. Then you say, “Actually, I don’t think our app explains it very well – this is how we had envisioned it working: [describe] – does that make sense to you?”

Maybe you were forced to add security features that were un-user-friendly. Then you say, “The reason it works that way is [explanation]; I know it isn’t an optimal user experience but here’s the benefit: [describe]

Maybe you were just wrong about what an intuitive user workflow would be. Then you say, “I agree, it is confusing.  Can you walk me through how you use it, so that I can better understand how we might make it better?”

When you start feeling defensive, it’s usually for a reason – you know something isn’t really quite right.  Use that feeling to learn something useful.

“It seems like I should be able to do…”

These are great because they reveal customers’ expectations and the way someone is thinking about the problem that your product solves.

This is what comes next, after customer development interviews and getting a first version of the product out there – making sure the experience makes sense.  You lay the groundwork for this with your initial research, but you won’t get everything right on the first try.  Certain workflows will just feel awkward.  Certain features — which no one remembered before — will suddenly emerge as glaringly absent.

There are times when a customer says “It seems like I should be able to…” and you disagree: you had a reason to not do it that way.  That doesn’t mean you should capitulate to what the customer wants; it also doesn’t mean you should ignore them.

Why you need to understand expectation mismatches

It’s a good opportunity for you to say “We’re not doing it that way … and here’s why.” In my experience, that usually kicks off a really useful conversation that helps me better understand customer needs.  (or, yes, about 5% of the time it leads to people swearing and ranting -but that’s about par for the internet)

They may be masking underlying problems. People who “expect” a product to work one way may have very valid reasons why it can’t work your way — the new way requires access to files they don’t have, requires them to install software but they’re on locked-down corporate workstations, etc.

It’s a silent dealbreaker. If something is blatantly broken, customers won’t hesitate to let you know.  But if it’s just not-quite-right, most people won’t complain – they’ll just stop using your product.  They may think “oh, it must just be me…probably other people wouldn’t mind this” — in fact, I’ve heard plenty of people say exactly that in usability testing sessions before.

Getting more “It seems like I should be able to do…” feedback

Whenever it makes sense, I try to ask “What would you have expected?” or “What did you think you’d be able to do?” in reply to more general support questions.

Also: [SHAMELESS PLUG AHEAD] I use KISSinsights to try to “catch” this kind of feedback when I think a page is “not quite right” (or even/especially if I think it’s awesome).

(All 3 of these questions are available in the free plan.)

Once you’ve found people for your interviews, you’re probably thinking, “Great, I can ask them if they’d use my product!”

Wrong.

OK, you’re thinking, “then what should I ask them?”

A better way to think about it is, what should I be learning from this interview?

It’s really important to understand the philosophy behind the customer development interview, particularly because it runs so counter to entrepreneurial instincts.  Be direct. Act. Get stuff done.  You ask questions, you get “70% good” information, you decide.  But when it comes to identifying a problem and potential solutions, the direct approach doesn’t really work.

Why? Because people are too polite to say ‘no’.  Because people can’t imagine technologies that don’t exist yet.  Because people overestimate how much effort they’re willing to put into something.  Because people think incremental, not disruptive.  Plenty of reasons.  Let’s accept this and move on.

So, what should I be learning from the customer development interview?

• How is your customer currently dealing with this task/problem?  (What solution/process are they using?)
• What do they like about their current solution/process?
• Is there some other solution/process you’ve tried in the past that was better or worse?
• What do they wish they could do that currently isn’t possible or practical?
• If they could do [answer to the above question], how would that make their lives better?
• Who is involved with this solution/process?  How long does it take?
• What is their state of mind when doing this task?  How busy/hurried/stressed/bored/frustrated? [note: learn this by watching their facial expressions and listening to their voice]
• What are they doing immediately before and after their current solution/process?
• How much time or money would they be willing to invest in a solution that made their lives easier?

The important thing about these questions is that they set up an environment where the customer is the “expert”.  They avoid yes/no answers, and give people the opportunity to tell a story – one that may trigger them to think of related problems they’re having, or may trigger more questions from you to ask later.

These questions are applicable for both consumer and enterprise products.  (I’ve used this question list on B2B internal tools, B2B2C consumer-facing apps, and B2C widgets.)

Can you give me an example?

These questions make a lot more sense when applied to a concrete example, so I’ll make one up: an online grocery shopping application. Your hypothesis is that busy households need a better way of making sure they don’t run out of things and don’t have to make a zillion trips to the market.  You’ve found customers to talk to – now you need to understand how they feel/behave when it comes to grocery shopping.

“Tell me about how your household handles grocery shopping…”

Whether they describe a detailed or haphazard process, this is your competition.  This is what you have to be substantially better than, in order to get customers to change their behavior.

“How is that process working for you?”

If you’re lucky, a customer may launch immediately into a rant on how they’re always running out of Cheerios or spending too much because they have to buy milk at the overpriced corner store.  If not, you may gently prompt them with triggers like “Do you generally have the ingredients you need to make dinner?” or “How much time do you spend shopping?”

Validation check: they might not care enough to change their current habits, even if they’re not 10o% optimal.

“Have you tried other approaches, like online grocery delivery or keeping a list on your iPhone?”

Customers who have tried other approaches = a good sign that this is enough of a problem that they’re motivated to fix it.

Validation check: Even if your customer thinks they spend too much time grocery shopping, if they’ve never tried any approach to fix this, then they don’t care enough to try your product. (On the other hand, if they have tried other things, you should try to learn why these other approaches didn’t help or were unsustainable.)

“If you could improve anything about your grocery shopping routine, what would it be?”

If customers don’t immediately have an idea, you could gently prompt with “spend less time, less money, have fresher/healthier foods on hand…?”

This question is going to prompt people to jump to solutions (like “I want a cost comparison tool”), rather than articulating their problems, so you need to immediately follow up with:

“If you had a cost comparison tool, how would that make your life easier?”

(basically, the 5-Whys approach)  You want to discover what’s at the root of this suggestion – is it more important that they pay the lowest prices, or do they want to cut down on trips to several different grocery stores, etc.

Validation check: if they really can’t articulate why this solution would make their lives better, it probably won’t.

“What people in your household buy groceries?”

These are the potential stakeholders of your solution.  This can also open up insights – if your customer says “I do all the shopping, but I wish my teenage kids could pitch in too”, that’s an area ripe for product exploration: how can we help division of labor? how can we shift simple tasks from the time-constrained/expensive resource to a “cheaper” one?

“What do you do immediately before you go grocery shopping?”

This is a great way to find ways to differentiate your product – to most people, grocery shopping starts when you walk through those electronic sliding doors.  But to your customer, it might start with asking your wife and kids what they need, making a list, looking up recipes online, or getting the baby changed and buckled into her carseat.

(Your product probably won’t solve for the baby thing, but it gives you insight into your customers’ state of mind – how busy they are, how stressed, do they only have one free hand to use…?)

“What do you do immediately after you go grocery shopping?”

Again, to your customer, grocery shopping isn’t over until the frozen foods are in the freezer and the cans in the cupboard.  This may reveal new product stakeholders: the 8- and 10-year old who never go to the grocery store, but who unload the car and put things away.

“Would you be willing to spend some money to get a cost comparison tool or other tools that would make your grocery shopping easier?”

If the customer says yes, suggest an amount (“say, $10 a month?”)

Validation check: If the customer says ‘no’, or says ‘yes’ in a hesitant way, they’re not going to use your product.

And finally…

After all these questions, feel free to ask about your specific solution and how interesting the customer finds it.  Give them the opportunity to ask questions of you.  You never know where inspiration will strike!

I asked myself, If I got this in my inbox, how likely would I be to reply to it? and immediately answered, Not very.

The problem with online tools is that they make it incredibly easy to solve problems, but in doing so, they sometimes lead you to solve the wrong problem efficiently.

With customer communication, the problem is not “contact customers and get it over with”.

Contacting your customer – whether it’s via email, phone, or carrier pigeon – is a means to an end: listening to what they have to say.

As some of you know, I’m the product manager for KISSmetrics, and the task at hand that day was to reach out to our earliest beta customers to check in – find out how they were using our product, nudge the ones who weren’t yet using it, and hopefully remind folks to send me some brutally honest feedback.

I had about 70 people to contact who fell neatly into two camps: those who were engaged and actively using the product, and those who had started the process but weren’t really engaged yet.

Clearly, the efficient solution was to prepare an email for each group and use Campaign Monitor to blast them out in a matter of moments.  Or was it?

I’ve sent out enough email newsletters and mass survey invitations to know what the average response rate is – 15% if you’re lucky, less than 10% normally.

And I had a hypothesis as to why. Bulk emails – no matter how carefully you craft the text – lack that sense of urgency.  It’s hard to imagine that there’s an actual human on the other end anxiously awaiting your reply. So even if you flag it for a response, you don’t feel any hurry to get back to the sender, and it gets buried in your inbox.

So I chose the deliberately inefficient method and got around a 40% response rate.

With some help from copy and paste, I sent out individual emails to each beta customer.  In taking the time to slow down, I remembered the people who’d had specific questions and added personalized notes.

Yes, it took longer – close to an entire day instead of the 5 minutes that a blast would’ve taken.  But it also brought in thoughtful questions and suggestions and kicked off longer conversations – the kind of feedback most startups wring their hands over not getting.

Is it practical to send out individual emails when your customer base is closer to 7,000 than 70?  Of course not.  But there’s no reason why you can’t select a random 10%, or even 1% of your customers to get the individual treatment.

Next time you’re thinking of sending out an email newsletter or a survey invitation, think about the kind of feedback you need.  Are you trying to answer “what?” or “which?”  If you need to answer “what”, be efficient: take the inefficient route.

P.S. Yes, I know — the best way to validate my hypothesis would have been to split-test and send half of my followups via mass email and the other half individually.  Maybe next time!

This scenario happens a lot with offshoring and virtual offices, and it’s why many companies have thrown up their hands and brought teams back in-house.

There, when confusion arises, you can immediately have a face-to-face conversation, maybe in front of a whiteboard. Discuss until you both get it. Simple, right?

Yesterday, I was having one of those “we are just not in sync” situations.  Except the coworker in question was 2,000 miles away. No face-to-face option, no whiteboard option.  Instead, we both had to battle through a long IM conversation, punctuated by “I sent you a quick sketch – is this closer to what you mean” interjections.

It was painful.  (Well, not so much for me – but almost certainly for him.)  But it forced us to be incredibly precise, and to keep confirming what the other was saying. “So you mean we’d default to this option and then continue to that screen?”  “No, let me rephrase…”

By the end of the conversation we were finally aligned.  But I couldn’t help but think that we would’ve given up sooner if we were both in the same location.  Why? Because proximity makes you lazy.

When you start getting frustrated, it’s easy to find an excuse to get out of there so you can go roll your eyes in private.  After all,  you both know that your colleague is just across the room.

But you may not be at your desk next time a question arises. Or, your coworker may feel self-conscious – jeez, she spent a half-hour explaining this to me, I feel stupid not “getting it”.  Maybe I should just figure it out on my own.

In the book Made to Stick, there’s an experiment described where one person tapped out a song in their head and the other had to guess what song it was.

The tappers got their message across 1 time in 40, but they thought they were getting their message across 1 time in 2… When they’re tapping, they can’t imagine what it’s like for the listeners to hear isolated taps rather than a song. This is the Curse of Knowledge. Once we know something, we find it hard to imagine what it was like not to know it.

I’d guess it has to do with proximity, too.  ‘He’s right there’, you think, ‘how on earth can he not get it?’

I’m certainly guilty of this.  If you look at my sent-mail folder, you’ll see some embarrassingly bad emails I’ve sent to local coworkers.  Things like “did you finish that doc for that meeting today?” Which doc? Which meeting? Could I be any more context-free?

We expect someone farther away to have more difficulty understanding what we mean.  So we make the kind of affordances that we ought to be doing all the time, for all people.

Next time you’re sending an email, explaining a concept, or scheduling a meeting, try pretending that your audience is a thousand miles away:

• Summarize what you’re trying to achieve
• State your assumptions (and ask that they be challenged if they’re wrong)
• Make a hypothesis (“I think this means X – is that correct?” is much more helpful than “I don’t understand this”)
• Restate the other person’s words (“So you’re saying Y and Z – is that correct?”)
• Finish with a clearly-stated bullet point summary of what you’ve agreed upon/decided

The extra five to ten minutes you invest in clarity will save you hours later on.

The concept of the “elevator pitch” may be a little bit cliched, but: let’s face it, people have limited attention spans.  If you want people to associate something with your product, your company, or you, it has to be short and easy to understand.

Here’s the updated pitch for this blog:

The experience IS your product.

To make customers happy and to be competitive, you’ve got to constantly work on describing, planning, iterating, researching, building, supporting, iterating, testing, and promoting it. This blog can help.

There are a lot of commas.  That’s deliberate.  People who have met me know that I get enthusiastic over all the details of building great products.  It’s not just requirements or markets, it’s understanding people and how they work and what they need.  I love talking about all of that.

Cindy’s tips for writing your pitch:

1) If you’re not sure how to articulate what your blog is about, run your URL through a tag-cloud generator.  It’ll show the words that come up over and over again. (For me: users, customer, launch, like)

2) Start by talking, not by typing.  You’ll use more natural language if you speak, wave your hands, etc.  It’s easier to take a good verbal snippet and make it a little more formal, than to try and force some life back into jargon-heavy marketing text.

Neither group distinguished between permission-based marketing and spam if they didn’t want the email.

Let me repeat that, because it’s so obvious that you may not have gotten it.  If your email is not useful, customers do not say “Oh, this email isn’t very useful, but I did sign up for it, so that’s okay.”

Are you ready for the best part?  Email Insider says:

“What’s clear from all of this is that many consumers don’t know how to use email properly-which is to say that they are a detriment to themselves and to email marketers.”

Wow.  Insulting your customers. (I had no idea there was a “right” and a “wrong” way to use email.)  Calling them a detriment to themselves and you.  (Yeah, in this economy, I can see how you’d have so many customers that they’d be a burden to you.)

Look – your customers not doing what you want is not their problem.  It’s your problem.

People drive demand, they pull out the wallets, they buy and use and pretty much insure you have a job (or don’t).   If people find your interface confusing, clarify it.  If people want to use your product in a different way, embrace it.

If people don’t find your product useful, BACK TO THE DRAWING BOARD.

The real lesson here is that customers don’t want email they don’t find useful.  The answer isn’t to educate them, it’s to educate yourself.  Create value, or get out of their inboxes.  Your customers know who the real “detriment” is.

Is this a surprise to anyone?

“Not only do blogs rank below newspapers and portals, they rank below wikis, direct mail, company email, and messageboard posts.” (People Don’t Trust Company Blogs)

The Forrester report goes on to encourage ways in which companies can make their blogs more relevant and more genuine, to try and reclaim some of that goodwill.

My thought is: not that many companies are interesting enough to merit a blog.

If you have an eloquent CEO or group product manager or other evangelist in-house, that’s not a problem.  But I’m guessing more often blog content is generated in the same way as it was at one of my past companies: “Intern, write a bunch of blog posts and date them every week or so.”  This is how company blogs end up with rehashed press releases or a description of the in-house poker tournament last month.

No, I don’t really care who went all-in, unless it was the CEO and that’s your way of telling us you’re going under next week.

Blogs imply timeliness and consistency in a way that other communication channels don’t.  People expect to see three, four, or six months between press releases.  If they see a blog with the last entry three months in the past, they assume it’s been abandoned.

So you need to post often.  And it has to be good.

Forrester says, please don’t stop blogging.  I say, don’t focus on the solution, focus on the problem you’re trying to solve.

Preview new functionality: not sure a blog post is the best way to communicate this, especially if you’re seeking feedback.  I’ve liked the experience of seeing a few screenshots or a quick demo in the context of a short survey.  It makes me like my consumer input may actually be listened to and acted upon.

Transparency, AKA Admitting You Were Wrong: if it was serious and affects a majority of customers, it deserves homepage billing.  If it affected only a portion of customers and you know the damage is contained, communication via email – quickly – may be your best bet.  You may not want unaffected people to know about an issue that will never affect them.

Community: This may be blogger heresy, but I think a blog is a terrible way to build community. As conversation goes, it’s incredibly one-sided.  Company talks, people can only really comment on the topics you raise.  At Yodlee, we had a bulletin board for our customers, and that gave us incredibly open commentary from our users. (Sometimes harsh and hard-to-stomach commentary, but you can’t shy away from listening.  The truth can be hard!)

A human face / Personality: This is great when it works, a la 37Signals.  You get a true sense of their personality, the way they think and listen and learn.  But I don’t think it’s always even appropriate.

My work with usability testing and financial software taught me that people don’t want to think of their bank as “human”.  Sure, they want to talk to a human when they have a problem.  But they’d prefer to think that anything involving their money is automatic, regulated, infallible.

You have to ask: does being “human” differentiate you?  Is it a reason why a customer would choose you over your competition?  If you’re a housekeeper or a tax consultant or you sell baby clothes, probably yes.  If you sell pacemakers or provide plumbing services… maybe not so much.  The one place where I think everyone needs to be “human” is their Careers section – but again, the solution to that problem is not a blog.

“If you want to be a thought leader and helper for your customers, and you blog frequently about those customers’ problems and solutions, then you can generate trust. This takes time and effort, but it will enhance your company’s reputation and it’s worth it.”

In the end, I don’t disagree with this.  But please, no more pocket queens and a third queen on the flop.

–

Update: Janet Fouts writes along similar lines, with Facebook May Not Be Right for Your Business.

If you’re so eager to communicate that you turn a conversation into a monologue, you’re not listening.

Recently I was interviewing a candidate who had enthusiastic, thoughtful answers … that weren’t quite relevant to our products/industry.  That was okay – I don’t expect interviewees to perfectly understand the company they’re coming into.  So I thought I’d clarify:

“For us, the single most important thing to communicate to customers is X.”

“Right!” said the interviewee enthusiastically.  “The most important thing for you communicate to customers is Y!”‘

I point this out only because – it’s obvious to see when other people aren’t listening.  It’s much harder to apply that same awareness to oneself.   And some weeks it feels like every conversation leads to another revision or another meeting.  Subconsciously, it’s tempting to put ourselves out of the reach of new information, because new information has a way of seriously screwing with those deadlines we have to meet.

So what’s a good product manager to do?

A couple habits I try to practice:

Ask at least one question before starting to present. Especially if you’re presenting to a customer or leading a meeting, it’s really easy to start talking and never really stop.   Your audience takes the cue from you and goes into passive mode.

Then listen.

Turn at least one statement into a question. Several months ago, I interrupted a customer check-in that was going smoothly to turn a statement — “So you’re happy with the product” — into a question — “What could the product be doing for you?”

Meanwhile, my subconscious was whispering: “Why would you do that?  They’re happy now.  Don’t give them any chance to think and be less than happy.  Get out of here as quickly as possible before your luck turns!” That subconscious, man.  Sometimes you just have to kick it back down.

Then listen some more.