Two different contracts with the customer
On-site chat and messaging apps look similar — a text box, a conversation, a person at the
other end — and they are not the same product. They make different promises, and choosing
between them is really choosing which promise you want to keep.
On-site chat is synchronous and contextual. The visitor is on a specific page, usually mid-task,
usually deciding something. The implied promise is that someone is there now, and the conversation
ends when the task does.
A messaging app is asynchronous and persistent. The customer sends a message from their own phone
and expects a reply eventually, in a thread that outlives the individual question. The promise is not
immediacy but continuity: the conversation is still there next week.
Most disappointment with either channel traces back to running one with the other's promise
— treating on-site chat as a place to leave messages nobody watches, or treating a messaging
inbox as though someone is standing by every minute.
Where on-site chat wins
Context is the decisive advantage. The visitor is on your pricing page, in your checkout, or
looking at a specific order, and the agent can see that. None of it has to be explained, and the
conversation starts from the problem rather than from establishing what the problem is about.
Intent is the second. Someone chatting from a product page is in a buying or blocked-right-now
state, which is when a fast answer is worth the most. On-site chat is where you catch the person who
would otherwise have closed the tab.
Reach is the third and it is underrated. There is nothing to install, no account to have, and no
network the customer must already be on. Anyone who can load your page can start a conversation.
Where a messaging app wins
Long-running relationships are the clear case. If a conversation naturally spans days — a
repair, a delivery, an ongoing account matter — a thread the customer can return to on their own
phone beats a session that ends when the tab closes.
Mobile-first audiences are the second, particularly in markets where a messaging app is the
default way people contact any business. In those markets, asking someone to come to your website to
talk to you is friction they will not always accept.
Notification reliability is the third. A message on a phone gets seen. A reply typed into a web
widget after the visitor has left may not be, unless you have a way to reach them by email.
The temptation is to solve the question by collecting every channel. It is rarely the right answer,
because the licence cost is the smallest part of the bill.
Every channel needs coverage during its own expected hours, and each one sets a different
expectation about what those hours are. Every channel needs its own tone, its own saved replies, and
its own place in the rota.
Worse, each additional channel fragments the customer's history. The same person appears as a web
visitor, a phone number and an email address, and the agent who picks up the third conversation has
none of the first two. The customer experiences that as being forgotten, which is exactly the problem
you added the channel to avoid.
A useful rule: do not add a channel you cannot staff to the standard it implies. One well-covered
channel beats three neglected ones, and neglect is far more visible on a messaging app, where the
unanswered message sits in the customer's pocket.
Two different contracts with the customer
|
On-site chat |
Messaging app |
| The promise |
Someone is here now |
A reply eventually, in a thread that outlives the question |
| Timing |
Synchronous |
Asynchronous and persistent |
| Context the agent can see |
The page, the checkout, the specific order |
Whatever the customer explains |
| What the customer needs first |
Nothing — anyone who can load your page can start |
The app, and an account on that network |
| When the conversation ends |
When the task does |
It does not — the thread is still there next week |
How to decide
Start from where your customers already are and what your conversations are actually like. If most
of your chats begin and end inside ten minutes and start from a product page, on-site chat is your
channel and a messaging app would mostly duplicate it.
If a large share of your conversations continue over days, or your audience contacts businesses by
message as a matter of course, that is a genuine case for an asynchronous channel — and the
practical version of it, for most teams, is email follow-up on top of on-site chat rather than a new
platform.
That combination covers the majority of what people actually want from a messaging app: the
conversation carries on, and the reply reaches them after they have left the page.
How MyLiveChat fits
MyLiveChat covers the on-site side of this and the follow-up that comes after it.
Website chat, offline capture and email-to-ticket follow-up run
in one queue today, so a conversation that starts in the widget can continue after the visitor
closes the tab without moving to a different tool or a different inbox.
Social messaging channels are not generally available today, and the product pages say so plainly
rather than implying otherwise. If a messaging app is genuinely central to how your customers contact
you, weigh that honestly — picking a channel strategy on a claim that has not shipped is how
teams end up migrating twice.
What to measure
Before adding any channel, measure how many of your current conversations actually need to
continue after the session ends. If the answer is small, your gap is coverage rather than channels.
After adding one, watch first response time per channel separately. A blended average hides a
neglected channel completely, and the neglected one is always the newest.
Track how often a customer has to repeat context that exists in another channel. That number is
the true cost of fragmentation, and it is the one that decides whether the second channel was worth
adding at all.
If you decide a messaging channel earns its place, the wiring is its own small project: see connecting WhatsApp, SMS and Messenger to your ticket queue for what each credential and switch on that screen actually controls.