You forgot to add some messages added by bots and the chatbox itself with links to FAQs and crap. It's not so straightforward to get to humans these days
Nice rant but, like those popups that ask for your email when you show signs of leaving the site , the proof is in the pudding - do these live chats increase conversions?
Agree. At one of my startups we hooked a $350k ARR enterprise customer over webchat. We also received some of the best early UX feedback through webchat.
Developers are somewhat different from the general population in that we like to read documentation rather than going to a person for help. But for almost everyone else having someone to answer a question immediately is the preferred experience.
How did you establish that the customer would not have contacted you if not for webchat? If i go to a company's website already intending to contact them i might type in a chat box before i email or call them, but they would have gotten my business regardless.
How did you establish that the customer would have contacted you if not for webchat?
In the parent's scenario, it seems likely that the buyer for that enterprise was simply browsing through the site, or perhaps looking at various options, when they found the parent's site, saw that there was a chat and saw the near immediate response, asked a few questions and eventually set up a meeting. To do that over email would have likely taken days.
To answer your question, they told us directly in our first sales meeting (we didn't close the deal over live chat lol). The director we were speaking to said something along the lines of "and to think we probably wouldn't be here if I hadn't written you in chat."
I'm not sure if your question is rhetorical, but I'd like to know this too. My observation is that what started as a decent idea was quickly hijacked by marketing departments and got corrupted so quickly that they've become an annoyance to me -- not unlike pop-ups, browser push notifications, and so forth. It's the "tinted car windows" of the web, immediately putting your site into a bad neighborhood.
Anecdotally, I've had far more success DM'ing companies on Twitter -- I don't think I've ever gotten a bot-like, automated response when taking that route, and I've had really specific questions answered quickly and thoroughly.
Where it has... profiles approaching me, with photos of females who wouldn't even respond to my "hi" in real life? You're are not going to earn commission on me sweetie.
Noooooo (but it will happen) - so we can have convincing but wrong content about company products!
Readers are commenting that chat increases "conversions" (I guess that means sales). I'd be interested to see what convincing but wrong chatbots do do conversions and to ongoing customer numbers.
Eh. It's good for B2B. If you're ordering technical stuff and need something specific adjusted, can be helpful to chat to someone that can connect you to a specialist quickly.
Eh, in my experience, I generally disagree. Why wait for an email response when you can get an answer then and there? Also, outside of online retail shopping, I've found live chat very useful.
Sure, get rid of the intrusive popups and chatbots which don't help. But keep the live chat. It's valuable to the customer.
In my experience the realtime chat agents appear pressured to clear issues immediately at the expense of correctness. Sending an async message gives the respondent time to research things or ask for help instead of making up answers on the fly.
I have a feeling the author does not have enough experience with e-commerce. Adaptation of livechat increased conversion rates and sales and it is visible in the reports. Return rates also go down with livechat. Especially women's products (hair, makeup, shoes) are benefiting from instant answers and change the conversion rates dramatically.
Having a human talk to you thru your screen is a game changer in e-commerce. I personally use amazon customer support chat myself very often.
It's the best way to do something you can't on the website. As example, I've had to return products that have said they were not returnable or beyond the return date, chat is the easiest way to talk to someone and get it resolved.
Yep, the author is thinking like a developer, not a consumer.
> If I need assistance, I will most likely do one of the following:
>
> Find a support e-mail.
> Go to a physical store.
> Ask a friend.
> See if I find an unintrusive live chat that I chose to enable.
This is a developer mindset: take the time to solve the puzzle. I get it, but the majority of people don't think that way.
Most people don't want to solve the puzzle. They want to buy the product and get on with their day. Live chat lets them do that, and by the numbers, it works.
A few weeks ago, I had Amazon accidentally send me someone else's package, who lives about five blocks away from me. Not even the same street name. But it reported my own package as having been delivered, which clearly was not the case.
I wrestled with that stupid chatbot for probably two hours. There was no option in its menu for this specific problem. It would only allow to report not receiving something, and the only option given to me was to wait a few more days, or report the wrong item had been delivered, in which case I needed to send it back. I tried calling the support phone number instead, but it was just the same chatbot but over voice, with the same options.
But this wasn't my problem. I had the right package. It just wasn't mine. It wasn't addressed to me. If I returned it, how would they know if it was me who sent it back?
I don't even remember how I finally found a phone number to call that connected me to a human, but I'm pretty sure I had to outside of Amazon to find it. As soon as I got connected to a human and explained what happened, they were more or less instantly able to help me, but it was incredibly frustrating not having the bot itself give me an option to talk to a human. I could only complete it's automated workflows.
Amusingly, the CSR told me to just keep this other guy's package and they'd send him another, even though I could have easily walked down the street and given it to him. Fine, I guess. It was a pressure washer attachment that actually does fit my pressure washer.
> I had the right package. It just wasn't mine. It wasn't addressed to me.
I think this assessment was the root of your problem. This wasn't what happened. What happened was two separate things and the mistake was tying them together.
You didn't get your package they claimed to have delivered. Full stop. Deal with that problem.
Second, you received a package addressed to someone else. Deal with that separately.
I think there's a difference between live _support_ chat and live _sales_ chat. If I'm seeking support I'll take live chat any day of the week.
I actually just complained about this yesterday- I do _not_ want to have your shitty version of clippy[0][1] flying around the screen, pulsing, and asking if I need help while I'm browsing your page. Stick it on the support/contact page and that's plenty good for me.
The availability of chat isn't what's wrong. It's the implementation. It should just be a static button in the header/footer or where ever. When I need it, I can find it, just like every other link.
The problem is that as a customer, it's hard to distinguish between the chats that are actual, human-will-respond-in-near-real-time chats, which are just a piss-poor FAQ bot that leads nowhere, which ones are a FAQ bot leading to a human, and which ones will ask me for an e-mail address (and if they do, which ones will spam me instead of contacting me in case a followup is needed).
Actual chat that's passively available on request and makes clear what it is is great. Anything that pops up, makes noise, or otherwise begs for engagement, or isn't "click here for human", is an anti-feature.
So the issue here is the automatic popup. If the issue is not about content (not everything is simple to find out), live chat is helpful. But it's only helpful if the support team is ready to help via email.
The kind of people it’s intended for, don’t know how to open a chat. Some people are afraid of clicking unknown icons. Some (different) people are afraid of starting a conversation.
On the Nespresso website, the chat button covers the Buy button partially. So I ask them “Please confirm my order”. Chats are almost always a horrible implementation.
Personally, I feel chat is too immediate and am easily irritated since whatever I need has me upset. I prefer an email and a fast response or a phone number to call.
In my experience, the usefulness of live chat features varies wildly.
If the live chat service connects me to a human who is able to resolve my query in a timely manner, I'll take that any day.
Phoning a large company can result in waiting in a call queue for a long time. My attention is on the phone call instead of anything else. A live chat window can be relatively ignored and checked once in a while.
Emailing a large company can result in waiting for a reply taking 3-5 business years. Live chat gets me a response relatively quickly.
A live chat service that does not provide an easy means of interacting with a human and which instead easily allows getting stuck in a bot loop is a great example of poor customer service and, for me, a great way of quickly finding out who to not do business with.
That said, a poor live chat service for a business that I absolutely have to contact (most recently when terminating internet service after moving house) is an exercise in frustration and annoyance that is hard to replicate in any other manner.
Exactly. Good chat implementations are great; meaningful automated information, fast handover to human operators while also providing them with the relevant information, and all that in the comfort of my browser window is a great experience.
Being stuck in a loop against what's essentially a buggy ELIZA-Clone however, or a system that somehow is incapable of authenticating me even though I've logged into my account on their website, is not.
From the other side of the fence - you can base your FAQs on actual questions (to the point of highlighting and bolding the most common) only to keep getting a significant amount of requests with that exact same question multiple times a day.
I can confirm that - at least in my experience - FAQ's are pre-made (and never or rarely updated), when a new service/site/whatever starts, it is logically nonsense that there is even one FAQ.
In theory everything should be clear from the info/documentation/manual, and when it is not so (as it often happens) and something is actually frequently asked, not only an entry in the FAQ list should be added (together with its FGA[0]) but the sheer fact that it is frequently asked should mean that the topic is not clear enough in the info/docs/etc and these should also actually be corrected/updated.
As a "disciplined" user (who actually did read both the FAQ's and the documentation) when you (manage to) contact (via mail/chat or phone) the assistance with a question/doubt there are usually three possibilities:
1) the assistant knows less than you on the topic and cannot answer properly
2) the assistant is competent and manages to answer your question, though with some difficulty (you made an original question)
3) the assistant is competent and answers your question easily because it has already been answered by him/her tens or hundreds of times (your question was not so original but never made it to the published FAQ's)
#3 is the clear sign of a failure in the way FAQ's are managed.
> Acronym for "frequently asked questions"; a list of answers to frequently asked questions that can be presented to a community (be it a forum, Usenet newsgroup, or software user base) so that the same questions need not be asked over and over again. In the entire history of their use, not one has ever been used for its intended purpose.
A lot of the people on HN might be very willing and able to search a FAQ to try and answer their own questions.
But you would not believe how many average people refuse to spend even 10 seconds trying to answer own question and insist on reaching out to a human even when the answer is very obviously readily available. Sometimes this makes sense; there's some scams out there and at least speaking to a native English-speaking person is reassuring when it comes to who you trust your money with. But sometimes some people are just miserable and want to annoy others.
What the chat should have is something like a chat interface that exposes the FAQ via question and answering, through natural language processing, basically like ChatGPT but it doesn't need to be that advanced.
If the bot can't find the answer, connect to a human. That way, the company's support burden is decreased while still being able to talk to a human if needed.
This could be an email, but of course the business doesn't want this as it would create a record of interaction. Whenever I had to use live chat it was a tactic to wear you down.
I just now chatted with Amazon, and not only do they not send (or ask to send) a transcript, they even clear the chat once the agent has left so you can't even save it yourself.
I can't imagine any other reason but "we don't want the customer to be able to have a record".
Before signing up to a new service, I tend to contact their support first to assess how difficult it is to contact them.
If I have to wait for an hour to get support, or navigate through some maze to reach them, or find it difficult to attempt to terminate my contract, I'll go with someone else.
When I really need help or something is wrong, that's not when I want to find out that it is almost impossible to reach them.
Note to self: Use ChatGPT to immediately respond to any support request that aren't from existing users so that potential customers have to perform the Turing test.
Plot twist: the chat bot is also running the Turing test. If the potential customer passes, they can sign up. Otherwise, “Contact us for pricing” and I don’t mean it as an anti-bot feature…
> A live chat window can be relatively ignored and checked once in a while.
Some live chat implementations have taken the liberty of requiring an input from your end every X minutes, otherwise the chat terminates. Support agents have literally told me "please wait a while while I look into this" followed by "are we still connected?" just a few minutes later.
However, you can't expect a company to tie up a support person for 48 hours on their screen and you get back to them 2 days later, that is ridiculous.
Where is that line drawn that is not too short and not too long? I don't know, but there has to be a line.
I do think it is my responsibility to continually check my screen to see if there's been an answer to my last message. There are two sides and both must be responsible actors. I think 5 minutes is plenty of time, for example. As the saying goes, "You snooze, you lose."
And it has happened to me and everyone else, of course. BUT, and here is the cool thing - when I get back on the chat, I just say that I was chatting and got distracted...but then the tech support person can go review the notes and quickly get up to speed on the conversation.
> Emailing a large company can result in waiting for a reply taking 3-5 business years. Live chat gets me a response relatively quickly.
I've had the same experience, and I don't understand why. In both cases, someone needs to read what you write and reply to it with a message. Why can't the just treat email just like live chat?
I dislike live chat because it forces me to wait for their responses (I'm looking at you, Amazon, minutes to type 10 words?) and dedicate time to it instead of putting all the information into an email and sending it and communicating asynchronously.
Agreed that sometimes they can be too nagging, but I always really appreciate being able to resolve issues in a live-but-async manner. It also seems that chat agents are usually more empowered than phone agents when it comes to resolving issues, and it's an added bonus that it's all in writing so everyone agrees what happened.
I spent 20 minutes in live chat with T-Mobile on my mobile device. Eventually a device reboot was required. I then spent 10 minutes establishing context, then 20 minutes in live chat with T-Mobile. Another restart was required. Repeat. Ad nauseum, or at least until I determined it was no longer worth my time to pursue the promotion I was supposedly eligible for.
I have never been pleased to see the chat bubble pop up on any site I've used. I do hope the people that use it enjoy it thoroughly, just to offset my disdain.
The problem isn't live chat. The problem is incorrectly implemented live chat.
I run a small saas and I can't count the number of times someone messaged an inquiry and I immediately got in touch with them. They were astounded by the quick response time, which is simply not possible with contact forms.
They were astounded by the quick response time, which is simply not possible with contact forms.
I'm intrigued. Why isn't it possible with a contact form? The mechanism for routing a message to you could be literally identical whether it's a chat box or a form. You could get the exact same notification on your system. It'd take you the same amount of time to respond. The only difference I can see is that what you reply with would go to the chat window in their browser instead of their inbox. Perceptually they may see that as faster I guess but it isn't really.
However...
I run a small saas and I can't count the number of times someone messaged an inquiry and I immediately got in touch with them.
What you presumably don't see, or at least ignore, is the number of times people didn't get an immediate response from you and were annoyed by it, or were annoyed by the chat option being there in the first place, or who didn't even see the chat option, or whose browser failed to load it, or... well... lots of things.
Your claim that users like it is simple survivorship bias - you're only focusing on the positive outcomes.
> I'm intrigued. Why isn't it possible with a contact form?
- E-mail always has a round-trip latency attached, even the best systems will have at least a minute, and you have to do insane amounts of work to keep up with "deliverability" as automated anti-spam systems can and will detect high frequency of incoming emails and downgrade your score, not to mention you have to deal with bullshit like contacting providers that downgraded or outright black-holed you.
- Uploading or sending attachments is a hit-and-miss, there still are providers limiting attachment size to 1MB (looking at you web.de) or with overzealous "security software".
- People will readily ignore the "do not reply below this line" and reply inline, which software like JIRA SD won't recognize or garble up.
- On mobile, people will have to switch between apps, which more often than not even on flagship devices leads to Android killing off the apps between switching for "memory efficiency" and 10+ seconds of load time after each switch.
Email certainly has its flaws, but none of that has any material impact on how quickly chucky123 can reply to a contact form rather than a chat window. A contact form on a website just sends a POST request to a server, and then that server notifies chucky123. The process could be identical to using chat in that respect (it often isn't, but we're talking about a specific instance rather than the general case.)
If chucky123 replies to the notification immediately via chat or via email the user should get pretty much the same experience. Suggesting this is a benefit of chat doesn't make much sense on a technical level. It isn't. It's just a benefit of replying to contact requests quickly.
If you're going to use email for customer interactions, then respect the conventions of email. I will reply below whatever line I choose. If your software doesn't support that, then answer the phone.
On point. Plus the fact that chat has conventions that closely related to how we talk naturally vs etiquette based emails means that you can drop a whole bunch of charade and just talk to your users naturally.
Another factor is not everyone is a hacker who reads their email after sending a form or understands how all this works in the first place. Most people have {e|g}mail because it’s that sign-in thing that their hacker relative set up for them.
Another big chunk of people doesn’t get that they can tap on a chat button, that’s why it flashes and pops up automatically or after a timeout of inactivity, which means a user is probably confused.
I wish there were levels of accessibility in the web tech. So you could specify not only whether a user can see or hear at all, but also select from few levels of computer knowledge, from beginner to confident to 30 years of software development, so everyone could get a reasonable experience.
Actually I wonder if this could be a growth area for Slack. E-commerce chat.
I had to chat with Verizon reps yesterday to set up my phone. They asked for my device ID or some other configuration numbers. I sarcastically put in 11023 <a lot of zeroes> 47781 (number is fake for this example).
And what appeared in the chat was 11023 47781. The chat app truncated anything between <>. Probably an attempt at XSS protection. But not the behavior I was expecting.
So if a company doesn’t have the capacity to do chat well, why not outsource it to a company that can? Make the integration easy, style the pop up to each company that uses it. Boom, new revenue stream.
If I'm browsing an SaaS and see a chat box that says someone is online. I'm much more likely to ask a question to the chat where I would otherwise just leave without trying to contact support in another way.
It seems to me that someone could just as reasonably (that is to say, not at all reasonably) argue:
> I don't like hunting for information when I'm shopping online, and would rather get a personal recommendation from a staff member. If you need to fill your website with loads of text, you probably need to do something about your live support staff.
Our individual personal preferences prove next to nothing about what a business "should" do.
Maybe it's true that these popups drive more people away than they convert on average (although I doubt it), but to claim that with any credibility you need to give some evidence.
Chats are cancer and useless. Search (not ads, not suggestions!) and category-aware filtering is what I want. For customer support ticketing system, email, phone.
A lot of the chat bubbles that I see flash in the corner, and make the title of the page flash as well; with lots of annoying bleeps and bloops.
I also find that sales is very trend driven. Currently they're trendy, but I hope that sales websites learn that you can't just stick things in people's faces all the time.
Fuck those chat bubbles that automatically open when I visit websites.
"Hi, Thank you for visiting. Please look around and let us know if you have any questions"
I'm gonna really reach out to you if I run into questions but don't automatically expand the chatbox and ruin my experience here. I vividly remember car dealership websites being the worst of them all. They all use the same software more or less and it is just designed to generate leads through coercing visitors and customers.
I wonder how many people just try to be chatty. "Hello! I'm so glad to hear from you! How is the weather where you are? How is the employer? Are you happy there?" Etc
Well, I guess this is what happens when a business operation is placed under the control of marketing.
It's a very old-fashioned attitude; back in the early noughties, many companies thought of the web as essentially an advertising billboard, so the website was placed under the marketing team. That's counterproductive, if the website is actually a core business operation.
It's pretty sad that even little startups these days are ending up doing it. They usually have two or three people "available" through chat but the chatbox punches you in the face even when you're just a visitor trying to check out what the fuck their startup does.
If they could, they'd trigger something on your phone or computer that causes a knife to stab you in the face the first time you open their site. But alas, their power is limited to rendering the site unusable and unreadable.
I just type "No thank you, I have no questions yet!" in those boxes. Then shortly after you will something like "<Support person Foo has entered the chat.> <Support person Foo has left the chat.>".
If everyone would do this then eventually the company would be forced to remove the initial chat bubble since their support is wasting 90% of their time looking at bullshit "No thank you!" answers.
> If everyone would do this then eventually the company would be forced to remove the initial chat bubble since their support is wasting 90% of their time looking at bullshit "No thank you!" answers.
or they'd get worse since someone would somehow figure out a way to prove it's "customer engagement" and therefore valuable.
Whenever it nags, I click on it and ask how to remove the chat.
Then I say it doesn’t work, because, clearly it doesn’t. At one point they give me a real person to talk to.
That’s what in-product char is for, right, resolving UX issues?
I’ve been looking for months for a time-tracking tool on MacOS and I never subscribe to any because I throw the ball when opening the pricing page and the chat opens. I’m so annoyed I insult them, I really need a time tracking tool for my team.
I've found adding a live chat to my online creation tool invaluable for getting a sense of what people find confusing or unintuitive. I change or update ux flows based on often-asked questions, and know if the change worked based on whether I still get those questions.
Chat is one thing. Obtrusive chat is another. Bot chat is a sort of pinnacle of shitness that eclipses all others.
When companies do chat and it actually helps the user, all good. When it's done entirely to help the company by reducing salaries for real staff / trying to prevent users getting in touch / promoting some feature or product the user didn't want in the first place... - then it's generally all bad.
Spectrum (the ISP) had terrible incompetent bots attempting to pose as humans last time i was forced to deal with them. They were even trained to claim they were human, but couldn't answer questions any person trivially knows the answer to. It's a true insult to the intelligence and time of their (unwilling) customers. The purpose is to waste peoples's time and grind them down in hopes they give up before even entering the phone tree.
I had a live chat widget on my startup's site for a while. What I found was that maybe 10% more people reached out, but the bigger difference was that about half of the contact that would have come through our email form (and not required an immediate response) was now coming through chat. This meant that we needed to respond immediately, which was a downside.
The most annoying thing though was when people would try to prove that I was actually a chatbot.
Any tips for how to definitively prove you're not a bot (shy of getting on a zoom call, which is what one user suggested!)?
I think you don't need to definitely prove you're human or pass some kind of adhoc turing test. You just need to convince them that whatever need they have can be taken care of via the chat widget.
If the need is to speak to a human, help them schedule a phone call with a sales rep / support rep / relevant party.
The hope was that more of the interactions would be ones that uniquely benefit from live chat. We found instead that it was just shifting our existing conversations from email to chat, typically with no benefit.
The problem isn't live chat, or nagging - it is a corporate policy structure for something complex (sites like eBay, Facebook, and Amazon are effectively running small municipal governments at least) combined with an incentive structure to get the person to tie up as little support time as possible, because that costs money.
Added in is the modern trend to try to make immensely complicated things like "global commerce" look like it takes two button clicks. Abstractions are good, but something, somewhere, should tell me how my inputs to this system affect what happens.
I think live chat is brilliant when I need it. But I partially agree with the article. Don't shove it it my face, that gives me the sleazy salesman feeling.
And I mean live chat. Me chatting with a bot is not a live chat.
What's going on recently with all the posts about "you don't need that" or "you don't want to be". Of course, I need a chat on my site - my whole business has been relying on this.
Only as long as the chat immediately connects you to a human and you dont have to click/write through 10 bullshit messages about nothing with some half-assed bot.
Yes this. If your thinking of incorporating live chat then make sure it's backed by humans or at least a way to get to human easily (if your AI bot fails to address the initial question). When you don't have a human available such as outside normal business hours just hide the chat feature, people can use your contact form.
IMHO this is the only acceptable method for having a chat feature on a site.
234 comments
[ 7.2 ms ] story [ 247 ms ] threadNo.
> An agent will be with you shortly!
> Hi, what can I help you with?
> Agent ended the session.
< I want to know about <something>
> An agent will be with you shortly!
> Please provide:
> your name
> your email
> your phone
No, thanks.
Developers are somewhat different from the general population in that we like to read documentation rather than going to a person for help. But for almost everyone else having someone to answer a question immediately is the preferred experience.
In the parent's scenario, it seems likely that the buyer for that enterprise was simply browsing through the site, or perhaps looking at various options, when they found the parent's site, saw that there was a chat and saw the near immediate response, asked a few questions and eventually set up a meeting. To do that over email would have likely taken days.
Source: I've had something like this happen too.
Anecdotally, I've had far more success DM'ing companies on Twitter -- I don't think I've ever gotten a bot-like, automated response when taking that route, and I've had really specific questions answered quickly and thoroughly.
I know if at least one case where it has...
Readers are commenting that chat increases "conversions" (I guess that means sales). I'd be interested to see what convincing but wrong chatbots do do conversions and to ongoing customer numbers.
Sure, get rid of the intrusive popups and chatbots which don't help. But keep the live chat. It's valuable to the customer.
Having a human talk to you thru your screen is a game changer in e-commerce. I personally use amazon customer support chat myself very often.
You answered that. Thanks.
For what?
> If I need assistance, I will most likely do one of the following:
>
> Find a support e-mail.
> Go to a physical store.
> Ask a friend.
> See if I find an unintrusive live chat that I chose to enable.
This is a developer mindset: take the time to solve the puzzle. I get it, but the majority of people don't think that way.
Most people don't want to solve the puzzle. They want to buy the product and get on with their day. Live chat lets them do that, and by the numbers, it works.
A few weeks ago, I had Amazon accidentally send me someone else's package, who lives about five blocks away from me. Not even the same street name. But it reported my own package as having been delivered, which clearly was not the case.
I wrestled with that stupid chatbot for probably two hours. There was no option in its menu for this specific problem. It would only allow to report not receiving something, and the only option given to me was to wait a few more days, or report the wrong item had been delivered, in which case I needed to send it back. I tried calling the support phone number instead, but it was just the same chatbot but over voice, with the same options.
But this wasn't my problem. I had the right package. It just wasn't mine. It wasn't addressed to me. If I returned it, how would they know if it was me who sent it back?
I don't even remember how I finally found a phone number to call that connected me to a human, but I'm pretty sure I had to outside of Amazon to find it. As soon as I got connected to a human and explained what happened, they were more or less instantly able to help me, but it was incredibly frustrating not having the bot itself give me an option to talk to a human. I could only complete it's automated workflows.
Amusingly, the CSR told me to just keep this other guy's package and they'd send him another, even though I could have easily walked down the street and given it to him. Fine, I guess. It was a pressure washer attachment that actually does fit my pressure washer.
I think this assessment was the root of your problem. This wasn't what happened. What happened was two separate things and the mistake was tying them together.
You didn't get your package they claimed to have delivered. Full stop. Deal with that problem.
Second, you received a package addressed to someone else. Deal with that separately.
If you need the package soon, that’s not great.
I actually just complained about this yesterday- I do _not_ want to have your shitty version of clippy[0][1] flying around the screen, pulsing, and asking if I need help while I'm browsing your page. Stick it on the support/contact page and that's plenty good for me.
[0]: https://media.welsh.cc/Mod27D
[1]: https://media.welsh.cc/e1tniz
Actual chat that's passively available on request and makes clear what it is is great. Anything that pops up, makes noise, or otherwise begs for engagement, or isn't "click here for human", is an anti-feature.
On the Nespresso website, the chat button covers the Buy button partially. So I ask them “Please confirm my order”. Chats are almost always a horrible implementation.
If the live chat service connects me to a human who is able to resolve my query in a timely manner, I'll take that any day.
Phoning a large company can result in waiting in a call queue for a long time. My attention is on the phone call instead of anything else. A live chat window can be relatively ignored and checked once in a while.
Emailing a large company can result in waiting for a reply taking 3-5 business years. Live chat gets me a response relatively quickly.
A live chat service that does not provide an easy means of interacting with a human and which instead easily allows getting stuck in a bot loop is a great example of poor customer service and, for me, a great way of quickly finding out who to not do business with.
That said, a poor live chat service for a business that I absolutely have to contact (most recently when terminating internet service after moving house) is an exercise in frustration and annoyance that is hard to replicate in any other manner.
Being stuck in a loop against what's essentially a buggy ELIZA-Clone however, or a system that somehow is incapable of authenticating me even though I've logged into my account on their website, is not.
Do you have examples?
When I think of FAQs, IIRC almost all the time it seems like they wrote it before release and never upgraded it with actual questions.
But in most cases I think it is both: FAQ isn't updated/relevant && people don't check it.
In theory everything should be clear from the info/documentation/manual, and when it is not so (as it often happens) and something is actually frequently asked, not only an entry in the FAQ list should be added (together with its FGA[0]) but the sheer fact that it is frequently asked should mean that the topic is not clear enough in the info/docs/etc and these should also actually be corrected/updated.
As a "disciplined" user (who actually did read both the FAQ's and the documentation) when you (manage to) contact (via mail/chat or phone) the assistance with a question/doubt there are usually three possibilities:
1) the assistant knows less than you on the topic and cannot answer properly
2) the assistant is competent and manages to answer your question, though with some difficulty (you made an original question)
3) the assistant is competent and answers your question easily because it has already been answered by him/her tens or hundreds of times (your question was not so original but never made it to the published FAQ's)
#3 is the clear sign of a failure in the way FAQ's are managed.
[0] http://jdebp.info/FGA/fga-not-faq.html
From the BonqQuest glossary https://www.jerkcity.com/glossary.html
But you would not believe how many average people refuse to spend even 10 seconds trying to answer own question and insist on reaching out to a human even when the answer is very obviously readily available. Sometimes this makes sense; there's some scams out there and at least speaking to a native English-speaking person is reassuring when it comes to who you trust your money with. But sometimes some people are just miserable and want to annoy others.
If the bot can't find the answer, connect to a human. That way, the company's support burden is decreased while still being able to talk to a human if needed.
I can't imagine any other reason but "we don't want the customer to be able to have a record".
If I have to wait for an hour to get support, or navigate through some maze to reach them, or find it difficult to attempt to terminate my contract, I'll go with someone else.
When I really need help or something is wrong, that's not when I want to find out that it is almost impossible to reach them.
Plot twist: the chat bot is also running the Turing test. If the potential customer passes, they can sign up. Otherwise, “Contact us for pricing” and I don’t mean it as an anti-bot feature…
Some live chat implementations have taken the liberty of requiring an input from your end every X minutes, otherwise the chat terminates. Support agents have literally told me "please wait a while while I look into this" followed by "are we still connected?" just a few minutes later.
However, you can't expect a company to tie up a support person for 48 hours on their screen and you get back to them 2 days later, that is ridiculous.
Where is that line drawn that is not too short and not too long? I don't know, but there has to be a line.
I do think it is my responsibility to continually check my screen to see if there's been an answer to my last message. There are two sides and both must be responsible actors. I think 5 minutes is plenty of time, for example. As the saying goes, "You snooze, you lose."
And it has happened to me and everyone else, of course. BUT, and here is the cool thing - when I get back on the chat, I just say that I was chatting and got distracted...but then the tech support person can go review the notes and quickly get up to speed on the conversation.
I love chat.
48 hours seems reasonable. Because that support person isn't tied up, they're answering other chats until you reply.
Otherwise, at least several times the time it took the agent to meaningfully respond.
I've had the same experience, and I don't understand why. In both cases, someone needs to read what you write and reply to it with a message. Why can't the just treat email just like live chat?
I dislike live chat because it forces me to wait for their responses (I'm looking at you, Amazon, minutes to type 10 words?) and dedicate time to it instead of putting all the information into an email and sending it and communicating asynchronously.
I run a small saas and I can't count the number of times someone messaged an inquiry and I immediately got in touch with them. They were astounded by the quick response time, which is simply not possible with contact forms.
I'm intrigued. Why isn't it possible with a contact form? The mechanism for routing a message to you could be literally identical whether it's a chat box or a form. You could get the exact same notification on your system. It'd take you the same amount of time to respond. The only difference I can see is that what you reply with would go to the chat window in their browser instead of their inbox. Perceptually they may see that as faster I guess but it isn't really.
However...
I run a small saas and I can't count the number of times someone messaged an inquiry and I immediately got in touch with them.
What you presumably don't see, or at least ignore, is the number of times people didn't get an immediate response from you and were annoyed by it, or were annoyed by the chat option being there in the first place, or who didn't even see the chat option, or whose browser failed to load it, or... well... lots of things.
Your claim that users like it is simple survivorship bias - you're only focusing on the positive outcomes.
- E-mail always has a round-trip latency attached, even the best systems will have at least a minute, and you have to do insane amounts of work to keep up with "deliverability" as automated anti-spam systems can and will detect high frequency of incoming emails and downgrade your score, not to mention you have to deal with bullshit like contacting providers that downgraded or outright black-holed you.
- Uploading or sending attachments is a hit-and-miss, there still are providers limiting attachment size to 1MB (looking at you web.de) or with overzealous "security software".
- People will readily ignore the "do not reply below this line" and reply inline, which software like JIRA SD won't recognize or garble up.
- On mobile, people will have to switch between apps, which more often than not even on flagship devices leads to Android killing off the apps between switching for "memory efficiency" and 10+ seconds of load time after each switch.
On-site chat has none of these problems.
If chucky123 replies to the notification immediately via chat or via email the user should get pretty much the same experience. Suggesting this is a benefit of chat doesn't make much sense on a technical level. It isn't. It's just a benefit of replying to contact requests quickly.
If you're going to use email for customer interactions, then respect the conventions of email. I will reply below whatever line I choose. If your software doesn't support that, then answer the phone.
Another big chunk of people doesn’t get that they can tap on a chat button, that’s why it flashes and pops up automatically or after a timeout of inactivity, which means a user is probably confused.
I wish there were levels of accessibility in the web tech. So you could specify not only whether a user can see or hear at all, but also select from few levels of computer knowledge, from beginner to confident to 30 years of software development, so everyone could get a reasonable experience.
I had to chat with Verizon reps yesterday to set up my phone. They asked for my device ID or some other configuration numbers. I sarcastically put in 11023 <a lot of zeroes> 47781 (number is fake for this example).
And what appeared in the chat was 11023 47781. The chat app truncated anything between <>. Probably an attempt at XSS protection. But not the behavior I was expecting.
So if a company doesn’t have the capacity to do chat well, why not outsource it to a company that can? Make the integration easy, style the pop up to each company that uses it. Boom, new revenue stream.
> I don't like hunting for information when I'm shopping online, and would rather get a personal recommendation from a staff member. If you need to fill your website with loads of text, you probably need to do something about your live support staff.
Our individual personal preferences prove next to nothing about what a business "should" do.
Maybe it's true that these popups drive more people away than they convert on average (although I doubt it), but to claim that with any credibility you need to give some evidence.
A lot of the chat bubbles that I see flash in the corner, and make the title of the page flash as well; with lots of annoying bleeps and bloops.
I also find that sales is very trend driven. Currently they're trendy, but I hope that sales websites learn that you can't just stick things in people's faces all the time.
I'm gonna really reach out to you if I run into questions but don't automatically expand the chatbox and ruin my experience here. I vividly remember car dealership websites being the worst of them all. They all use the same software more or less and it is just designed to generate leads through coercing visitors and customers.
Also, you usually have to close the "subcribe to our newsletter" popup first, to even see the chat window, covering the content you're interested in.
Then you scroll half a screen down, and a new popup with a survey, asking you how much you like their page.
Basically, they actively try to discourage you from seeing the content you actually want to see on the page.
It's a very old-fashioned attitude; back in the early noughties, many companies thought of the web as essentially an advertising billboard, so the website was placed under the marketing team. That's counterproductive, if the website is actually a core business operation.
If everyone would do this then eventually the company would be forced to remove the initial chat bubble since their support is wasting 90% of their time looking at bullshit "No thank you!" answers.
or they'd get worse since someone would somehow figure out a way to prove it's "customer engagement" and therefore valuable.
Then I say it doesn’t work, because, clearly it doesn’t. At one point they give me a real person to talk to.
That’s what in-product char is for, right, resolving UX issues?
I’ve been looking for months for a time-tracking tool on MacOS and I never subscribe to any because I throw the ball when opening the pricing page and the chat opens. I’m so annoyed I insult them, I really need a time tracking tool for my team.
0: https://www.intercom.com/help/en/articles/5053699-how-do-i-s...
When companies do chat and it actually helps the user, all good. When it's done entirely to help the company by reducing salaries for real staff / trying to prevent users getting in touch / promoting some feature or product the user didn't want in the first place... - then it's generally all bad.
The most annoying thing though was when people would try to prove that I was actually a chatbot.
Any tips for how to definitively prove you're not a bot (shy of getting on a zoom call, which is what one user suggested!)?
If the need is to speak to a human, help them schedule a phone call with a sales rep / support rep / relevant party.
Isn’t the whole point of having chat on your site?
Added in is the modern trend to try to make immensely complicated things like "global commerce" look like it takes two button clicks. Abstractions are good, but something, somewhere, should tell me how my inputs to this system affect what happens.
And I mean live chat. Me chatting with a bot is not a live chat.
IMHO this is the only acceptable method for having a chat feature on a site.