Running Surveys

[00:00:00]
>> Most of the time I fill in those gaps just for the survey, right? I'm a great believer in surveys and I use PollFish all the time for this because I can get results really quickly and I can define exactly who my audience is. I can say I want Welsh people who pick leeks on a Tuesday.

[00:00:20]
Maybe not that specific, but I can pick the kind of people I want, and I just take those gaps, those things that I don't feel I know, turn them into questions, shove that out, and get people to respond. If I've got the money to do that, right? If I don't have the money to do that, I mean, normally, I like to send out survey to like 130 people, seems to be a number I've randomly settled on because it's a reasonably big number.

[00:00:48]
But, of course, when you go into that number of people and if I have a lot of questions, that could be costing me $300, right? Which I don't always have, right? So if I don't have that, either, I'll get the client to send an email out to existing clients if I can, right, to their customers, in which case it's free.

[00:01:09]
Or failing that, I might just test with ten people, something is better than nothing, remember, right? A couple of tips if you ever do run a survey, cuz people love to pick other people's surveys apart. So here's a bit of advice to maybe avoid getting your surveys picked apart.

[00:01:31]
Generally, keep them short, people abandon longer surveys. And your stakeholders, they hear you're running a survey, everyone wants to tag a question on, right? So just don't tell them you're running a survey, [LAUGH] keep it quiet, keep it stealth, get it out the door, don't let other people interfere with it, keep it short, right?

[00:01:52]
Ask actionable questions. So ask yourself what you can do with the answer to that question, right? If you can't actually act on it, probably don't ask it, right? So people love to ask questions that they want to know. I feel I need to know how many times people have used the service before, right?

[00:02:17]
I've just made that up. But what are you gonna do with that knowledge, right? Let's say people have used it never before, or they've used it five times before or ten times before, how does that make a difference to what you do, right? Well, maybe it does, maybe if you've got a lot of power users and not many new users, you wanna build your application more for power users.

[00:02:43]
But unless you know what difference it would make, what action you would take based on it, probably, you don't wanna ask the question. Do you know what I mean? We spend so much time waste, so much time doing user research that we don't ever do anything with, so don't do it.

[00:02:58]
I tend to favor closed questions. So, reviewing all the answers to open-ended questions can be horribly time consuming, right? So if you ask a multi-choice question, you can instantly see which one comes out on top. If you have that same question as an open-ended question, you've now got go through 130 results trying to pick out trends.

[00:03:22]
That said, it's becoming a little bit easier these days with AI cuz you can just take the whole list, dump it into AI and say, find me trends, [LAUGH] and summarize this for me, but it's not as good as closed to questions. So I tend to, it's just cuz it's easier.

[00:03:38]
And then like I said earlier, check your wording, it's easy for there to be biases or poor choices of wording, and AI can help with that. You might meet some barriers when it comes to surveying. So I just wanted to kind of flag those with you. Because it sounds like it would be really easy just to throw survey together and get out the door, but nothing's ever easy, especially if you work for a large company.

[00:04:00]
Data protection is the big one. And so I always recommend, keep your surveys anonymous, right? Never ask people for their email address or their personal information. The minute you do, you're gonna run headlong into compliance issues if you're a larger organization, and everything will go into a hole.

[00:04:21]
So just kind of avoid that if you can. The other one is is survey fatigue. So, often the big thing I come up with cuz I like to ask users stuff a lot, and I get hit with this problem of, well, we don't like to send emails out to everybody all the time, it annoys them, right?

[00:04:45]
And so what I would advise you, if you don't have the ability to directly email people because you work for a slightly larger organization that got rules about this stuff. Then what you say is, I wanna send out one email, right, asking people if they would be willing to opt in to a user research or a user consultancy group, or whatever you wanna call it, right?

[00:05:11]
And then you get people to opt in to that. Now, you then end up owning that list, which means you can email them from that point on, because they've given you permission to email them, that you're not annoying the client base. You're only annoying the people that have said, I'm willing to be annoyed, right?

[00:05:30]
And you are the one that can send the messages out, so it saves a lot of time and a lot of hassle. But of course, that only really applies if you work for slightly bigger customers that have rules and regulations about this stuff. The final problem that you'll come across is poor participation, right?

[00:05:48]
Especially, this is more of an issue if you work for smaller clients, right? So you might get away with the data protection thing or the survey thing, but this will come back to bite you. So, this is why I consider using a platform like PollFish, which can recruit participants for you.

[00:06:05]
So this will get you a long way surveying, the only downside of surveying is you don't get to meet people and interact with people.