What I hear you saying is...

I stayed at "THE Hotel" last week in Vegas. Pretentious name aside, I fell in love with the place. It's perfect for a Vegas-hater like me, because the place is clean, cheap, modern and it doesn't have that casino stench that pervades every other hotel on the strip. I especially like that there is no casino in the lobby, but that one is accessible just a few steps away in the Mandalay Bay.

I'd been thinking good thoughts about this place for the last few days, and then checked my inbox to find that they'd sent me an online survey. 99.9% of the time I delete these, but given my recent experience I thought what the heck -- I'd help them out with a response.

Why are online surveys always such a dreadful experience?

There are at least a couple of big problems that I've experienced:

1) Too many questions, too little focus

The survey formats and questions are designed to solicit average responses from average customers -- not the passionate folk that have opinions (they are bored after the 5th question and drop off). Moreover, the "Strongly Agree/Disagree" multiple-choice format lures marketers into thinking that -- because they are so easy and quick to answer -- they can ask for the moon from their responders (e.g. 100+ questions).

So the only people that complete the surveys are those that enjoy the process -- which equals meaningless data from meaningless people. That is, the scores become diluted to the point that there are few data points worth communicating from the results. I used to see this all the time with Big Co. annual employee surveys. ("The overall employee satisfaction dropped from 2.8 to 2.7. What is the cause???")

2) Marketers don't understand or ignore available technology

Most marketers aren't taking advantage of obvious technology that would eliminate the need for many many of their questions. For example, if I'm entering their survey via a link from a mail they sent, why do they then ask me the dates of my stay, since they should be able to correlate this data with my mail account? (e.g. user "blake@mail.com" stayed here on dates X) Or, regardless of state, if on the very first question I've marked my stays as "Awesome" or "Horrendous", why march me through 50 screens of BS, when presenting me with a simple text-entry box on the next screen would probably elicit significantly richer, albeit subjective, feedback? At least they should take advantage of the flexibility in form-variables that you get with an online experience.

THE survey from THE Hotel was so annoying that I closed it after the 5th question out of boredom and irritation. Then, inspired to write this post, I tried to re-open the survey to cull some choice banality for examples. No dice -- Here is the message:

The survey you are attempting to access is still currently active on our servers. This may occur if there was a disruption in your connection to our servers and the survey has not yet been completed.

Please wait 60 minutes before attempting to re-enter the survey and use the URL link and password in your email invitation. When you re-enter the survey, it should resume at the point where your connection was lost.

Ermmmm. Not likely.

I'm actually surprised by this. Vegas is infamous for cutting edge analytics in their customer loyalty and retention. You'd think they'd be able to get the art of the survey right. You'd imagine they'd be reluctant to abuse the Permission Marketing model, too.

In contrast, they could learn a lesson from Eric Ries -- He recently ran a survey on his blog the other day with one required question: "How likely are you to recommend [this blog] to a friend or colleague?" That's it. (And incidentally, yes, I would recommend it). Every other question was gravy, but still thoughtful as well as 10x shorter than the dross I received from THE Hotel.

How could they improve? I like Seth Godin's post about this a few weeks back - abridged here for your amusement:

1. Don't ask a question unless you truly care about the answer.

2. Every question you ask changes the way your users think.

3. Make it easy for the user to bail.

4. Make the questions entertaining.

5. Shake up the format.