The question is about having customer survey or feedback survey on home page of the site. The said site, is used car buying and selling car.
In your opinion, should we have it to get feedback and in order to enhance the user experience?
User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It only takes a minute to sign up.Sign up to join this community
Consider what the purpose is of a survey, what value is it actually bringing to the end user? When are they going to hit the survey in their buying process? It's totally putting the needs of you as a researcher before the needs of the customer.
Surveys are tremendously easy to get wrong, hard to interpret well, and generally difficult to extract useful data from unless you really know what you're doing. Erika Hall, author of "Just Enough Research", wrote this fantastic Medium article about why surveys feel correct but generally don't give you usable data.
So in short, you can put a survey on your homepage, but you have to ask yourself what sort of data you're actually getting and if it's a useful reflection of reality. I say it likely wouldn't be.
Yes, it is very good practice and it's at the core of UX - getting feedback from users. However, it is important how exactly you will do it.
Try to avoid popup surveys. They are annoying and can cause your visitors to leave and never return. Instead include subtle link for feedback in the footer or sidebar so users can find it anytime and report issues/recommendations. For example, look at the bottom right corner of Google translate. You can also give them something valuable in exchange for a feedback. For example, if they fill up the survey you can give them free featured ad on your site, or unlock some paid features, etc.
If you are going to use popups don't display them immediately when the page is loaded because users will most likely ignore it. Also, there is no point to ask the users something before they've seen the website. Instead wait for the better moment - when they are going to leave your page, or when they're scrolling and reached the end of the page.
Don't ask too many questions. Users don't like to waste time filling up questionnaires. Ask 1-3 questions at most. They should all be carefully thought.
Open questions bring more value than closed ones.
It is important after you get the same feedback from couple of users to act and improve your website. You can also, incentivize users to give you feedback, as this information is much more valuable than the incentive monetary value.
My opinion is that if you really do need to make a survey, you should do it when the user is already somewhat familiar with your site. Perhaps after a task to which you know the user will feel at easy (after a purchase for example when the "buy or not to buy" is in the past). I say this because an online survey appearing at the homepage (somehow forcing you to at least perceive it) is invasive (I say this as an user, I do not have a reference on this). Also its likely that only some persons will respond to the survey (usually more patient, and as so you get a biased sample).
The reason why I mention the necessity or not to do the survey is that you can actually infer a lot of information from an indirect approach. Some sites mention some possible metrics (by sector). A few examples are:
You will often find mentions to online surveys as a way of measuring UX quality, and its certainly a possibility, but perhaps one that should be avoided until all other options (that do not rely on "bothering" the user) are used.
Notice, however, that there are someways of obtaining direct information from the user without any actual deliberate survey happening. This needs a careful approach, and likely some time used only for testing (since it implies some branding experimentation), but some possibilities are:
You can also provide a banner for a person to voluntarily do a survey (but once again its likely you'll get a biased sample). If you do decide to opt by the online survey check the web for tested examples for this kind of surveys.
I would agree that it is not uncommon and it can garner some good results, though I would not agree that it is "good practice"; a survey is good for you, not necessarily for your users. My $0.02 is that if you are going to do it, three important things:
Do not obstruct any of the user's tasks! The user is on your site to learn or accomplish something, not to take your survey. If they want to, great! But if not, don't get in their way - that will only rob you of good will.
Make it as concise as possible. A variation of the above rationale: if you want their input, take as little of their time as possible. This is one of the reasons things like NPS and CSAT are so popular: they are a single question, easy to administer, easy to score, easy to take for the user. You want to be considerate at that point in their journey - they are there for a reason and they probably don't want to take a lot of time, even if they are willing to give you feedback.
Consider and craft your questions well. This is the core of all user research, of course... if you don't ask the right questions, you won't get the right answers. Also make sure that the questions you ask are scoped appropriately to the context in which you are asking them.
Above all, do no harm... you can always end your mini-survey with something like "Thank you for your help. Your input is very valuable to us... if you'd be willing to speak with us further, please enter your email here."