Mike and Boris – co-founders of the popular quiz maker Riddle.com answer their favorite customer questions about how to build successful quizzes and personality tests. You’ll learn everything from the perfect quiz length and what questions to ask to crafting the ideal lead form – and everything in-between.
Hi there, my name is Mike – I’m one of the co-founders of Riddle.com and in this episode of The Quiz Makers, I am going to be interviewing and chatting with my co-founder and our CEO, Boris Pfeiffer.
Hey there, Mike.
Hey, Boris. Welcome back. So in this episode, Boris, you and I were chatting that ever since we started The Quiz Makers podcast, we get a lot of questions from people who are interested in using quizzes for marketing. But often, they’re often some of the basic fundamental best practices
I’d be happy to answer these and they’re not just feedback to our podcast. Mike and I also man our tech support chat on Riddle.com – it’s an interesting fact that despite having thousands and thousands of users, we’ve never hired a customer support team. We believe that if the founders and the lead engineers help in customer support, you end up getting a much better product. It’s been hard on us, but I think it helps, Mike.
Absolutely. Sure, sometimes we might be at the pub having a beer – and hear the support chat ‘ding’ on our smartphones, we quickly answer it. So it does distract us but the amount of positive feedback we got from users and also just great questions that lead to “Oh, these are features we should add.”
Anyways, that’s enough of Riddle. Let’s go talk about quizzes and marketing and all that good stuff.
So I’m just going to work through our list of the top 10 questions. One of the most common questions I’ve seen is “I want to make a quiz for marketing and I want to collect lead – what’s the ideal number of questions to have in a quiz?”
That’s pretty easy to determine. It depends on how complicated these questions are and how easy to answer for one, because you want to keep the quiz under two and a half or three minutes to answer. That’s about the most time people have attention for on the internet these days.
One caveat – if you’re doing a scientific research type of quiz, such as doing a medical quiz like “Do you suffer from anxiety?” People may be much more interested in that result and be willing to spend 10 or 20 minutes.
But for light-hearted quizzes – “What city should you live in?”, a good guideline around 10 questions is an ideal number, plus or minus. If you’re doing a personality test, sometimes you need a few more to score properly. But definitely try to stay under 15 is my guideline. We’ve seen much longer quizzes, but we don’t think they work that well.
No, it’s true. I would actually would say eight to 10. And a good way to test this is once you finished your quiz, just ask your Aunt Sally or your Uncle Bob to take the quiz and just time them (because if you take it again, you’ll know the questions). You will be much faster, you want someone completely new to the quiz to find out “How long did that take?”
And if they come back and say, “Oh, it took me about six minutes, then you know that you might want to shorten it down.
Okay, that leads into the next question. We’ve covered questions. Now how many answers? Should you have per question?
I would say four answer options is what people expect to have. There’s an additional benefit in having four if you use images, for example, it makes a nice 2×2 grid for the answer options. But four is what you see in most common multiple choice tests and quizzes – one correct and three wrong answers. So I would always try to stick to four if possible.
Yes, I absolutely agree. And there’s another benefit as well. Generally, 50 to 60% of your quiz takers will be on a smartphone. The fewer answer options you have, the more condensed your quiz will be. It’ll stay nicely on one screen.
Contrast that with adding six to eight to 10 answer options. Your quiz takers might need to scroll it.
Okay, next question. So should we use images in a quiz? Or should we stick to text only questions and answers?
I would always try to mix it up, have image questions followed by text questions, and then throw in another image question – it makes it a lot more interesting than just text. Sure, it may not be suitable for all quiz types. But if you can, in your eight to 10 questions, try to have at least one or two image based questions in there.
Definitely. And I would actually probably go 50% text / 50% image questions. For the image questions, maybe add a couple of GIFs or a video clip – just something to mix it up and add some energy and animation beyond just static images. But again, it depends on the subject. If it’s a very serious topic, you might want to avoid that.
Okay, this is a very popular question we get. How do you write a good quiz title?
This is probably the hardest thing to do when writing a quiz. If you’re doing this right, you’re going to spend as much time finding a good title as writing the rest of the quiz.
And the title is the most important thing – because it’s what you see when people shared a quiz on social media. It’s on the page where you embed the quiz. It gets people to click and start the quiz. So you have to get them hooked into the quiz.
It’s like the cover of a book. You know, when you’re looking at a bookstore, what jumps out at you? We’re not supposed to judge books by their covers, but we all do. So it needs to have a good image and a great title.
One other thing with titles – make them personal. So, for example, use the word you in the title.
If you’re doing quizzes or by lists, try to use odd numbers. There’s been lots of experiments and white papers written that odd numbers work better than even numbers. So instead of saying “Which of these six countries should you spend the rest of your life?”, use seven.
Or explore – you know, the “Seven most popular holiday vacation spots you can’t miss” works a lot better than explore the “Six most popular holiday spots”.
I honestly can’t explain the science behind it. But it’s common knowledge.
Now, that works. The other one – people have a tendency to say top 10s – like “Top 10 reasons to do this”. And I think people just kind of tune out.
But using odd numbers – let’s take “7, 9, 13 things you didn’t know about the Apple iPhone”. That just seems a bit more genuine and people go wow, that must be really interesting.
As we mentioned, don’t underestimate the power of the word ‘you’. Instead of “The dream city quiz”, you could go with “What city should you live in?”
That’s not it’s still not a great title, but just adding the word you makes it more personal.
And there’s one other trick. Challenge your audience. So instead of just saying, oh, here’s “The Harry Potter quiz”. Imagine we add the word you -“How much do you know about Harry Potter?”
That’s okay. But then you can also say “How much do you actually…” or “How much do you really know about Harry Potter?”
Just that one emotionally charged word, again, has a big impact. And it’s human psychology. Who knows what’s going on?
The last thing I would say is keep everything short. I think under 60 to 65 characters is your ideal quiz title target length.
Good advice – you probably see this in a lot of ads for video games on your phone. Underneath the “How much do you actually know about Harry Potter?” title – you could put a little byline saying “80% of quiz takers fail to get 100%”… something like that. Or “only 2% of quiz takers answer everything correctly”. You see this in a lot of in game apps where they said there’s a puzzle, and hardly anyone can solve this.
Yeah, it challenges people who say, “Well, I’m gonna be one of these 2% that gets everything correct.” And they’ll share it because they’re so happy.
Perfect. Okay, next.
So if you are using quizzes for collecting leads, and potential customers (and they are brilliant at that)… we’ve talked about that ad nauseum.
How many fields should you put into a lead generation form?
As few as you possibly can is always the best answer. Don’t go overboard.
Again, it depends on the type of quiz of course. And I think we’ll talk about this in the follow up question about what are the best quiz types for lead generation. But the more involved your users are by your quiz, the more fields you can probably ask for.
But ideally, if you can get by with just name and email, just ask for name and email. And remember, if you like you can collect and store all the quiz data along with the form. If you’re smart and asking the questions, maybe you can find out two very smart questions, things that you otherwise would put into the form so you don’t have to add many fields.
For example, let’s say for whatever you do in marketing, it’s really important that you know if someone’s under 20 or over 50, and those are the two brackets you really care about. Now, you could easily put a drop down into the lead form, “How old are you?” But people may not really want to answer that.
Or you could also have a question in the quiz like “Have you ever used TikTok?”
The chances that someone over 60 says yes are very low, while someone under 20. Very high. So you could use that question to determine the age range. And there may be others – whatever the data point, you can probably come up with a good quiz question to find out things.
That’s perfect. And one thing to avoid if you’re working for any medium to large company. If you are collaborating with other departments, they will often want to add lots of questions. So legal might want to put in five checkboxes saying “Yes, I agree to this.” and “I agree to that.”
Marketing might want to have an extra five fields for consumer research. Try and fight that as much as you can, because with each extra field, your conversion rates will drop dramatically.
Okay, let’s transition to something you mentioned it already. But speaking of leads, what is the best type of quizzes for generating leads?
The highest conversion rates will always be In quizzes that get people the most involved – these are personality tests. Because the personality test is all about you. You don’t have to prove you know anything like you do in a quiz. You do the test because you want to find out something about yourself. Like the “What kind of dog are you?” personality test we talked about that in the one of the previous episodes?
You’re not doing this to tell someone else what type of dog you are, you want to find out for yourself. And because you’re so involved, filling out that form, (which is just another question) after a couple minutes of quiz taking has such a much higher chance to be filled out.
If you do a quiz, like the one you mentioned Mike, that challenge you like “How much you really know about Harry Potter?”, they may also have a better chance than something light and fluffy that doesn’t really challenge you.
That would be my suggestion. Go for conversions with personality tests or in really challenging and difficult quizzes, where people really want to know how they how they ended up.
And you could promise them on the lead form that you’ll send them all the answer explanations and a follow up email (which is something our quiz maker supports), then you will also have a good chance.
Yes, because then people are saying, okay, I’ve taken this quiz, I’m going to get something of value. I’m going get all the quiz answers.
That’s awesome. Especially if it’s a really challenging quiz.
Okay, let’s transition slightly. So quizzes and personality tests are some of the best converting formats. One question we are often asked is, “Well, what about opinion polls?”
Because many publishers or businesses will use opinion polls in their content just to get people’s opinions, like the name says.
But are they good for gathering leads?
To an extent- the people there are not that involved. So your CTA (call to action) needs to be super strong.
Why should someone fill out the form? Maybe promising them to send them the poll results after the poll is over? If it’s a very engaging poll, that could be interesting. It would have to be a poll that really captures their interest and where they totally want to know what other people think. So let’s say you’re, you’re running a crypto exchange or currency exchange, and you have professional traders, and you’re asking for opinions: “What do you think? Where do you think Bitcoin is going to be a month from now?”
As a trader, I would be super interested in hearing what the other thousands of traders that take the poll think – it would give me a good snapshot. And yes, I would leave my email if you promise to send me the poll results from another thousand traders. But if it’s something “Do you think it’s going to rain tonight?” – that may be too light and fluffy. Sure, people can click on it if there’s a soccer game going on that everyone in my city wants to go to. But I’m not going to be interested in getting an email with these results, because I can just wait for it tonight, and I’ll see if it rains or not.
The one benefit of polls, though, is that they’re so easy to create. If you use polls in a lot of your longer form content, even if the polls don’t convert as well, they’re so easy to make that, even if you only get five or 10%, you can pop polls in your articles. That percentage is actually pretty low for us, even though it’s very good for the online sector. You know, that’s still not bad.
So polls do have a place in lead generation, as you mentioned.
Next… when people are creating quizzes, after writing six to eight questions, when they get to the end, they need to assign personality or overall quiz results. “Hey, you’re this” or “Wow, you got x out of y questions right”… how many different quiz results do you recommend?
I think four is again a really good number, as with the answer options. If you just do two, such as “You did really well” or “You did really bad”. That’s often too easy, especially with 10 questions, people will too easily fall into a low category. Three is kind of weird to set it up in thirds and four, you know, you get the top 25%, the bottom 25% and then the medium, high, medium, low people and quiz.
So for quizzes for personality tests, I like four to five. Don’t go over that because remember, your personality tests assign points to a personality based on the answer option chosen. You will need to have enough questions and answer options so every personality has a chance to get enough points to win.
If you do more than four or five, you are going to then have to write 20 questions with four or five answer options and the quiz will be horrible for most light-hearted or entertaining test topics. And you can work off the OCEAN personality types which is really five results. But as a rule, you can go with a kind of introverted result, we have the extroverted result, and now you have two extremes, you kind of do two in the middle – you’ll do fine.
Okay, now, that’s perfect. And the other thing about four, you know, four to five quiz results. There’s enough variety that when people share it on social media, their friends aren’t all going to get the same results. So there’s enough variety to say, “Oh, I’m this” or “You’re that”.
If everyone gets very similar results, that’s going to kill your virality. And people are just not going to be as impressed with your quiz.
We actually had two more questions, but because we like to keep our podcasts quite short and punchy and to the point, we’re going to hold off on those, we might bring this back in another Q&A session.
If you have any questions of your own, please let us know on Twitter. And because we’re online so much at Riddle.com you can also just go to Riddle and ask away on our support chat, we’re always happy to answer any questions.
Thank you very much, Boris, for your time, as always. Looking forward to pick this up with our next episode.