Tawk.to Is Free Until You Want Your Own Logo Back

The Little Badge In The Corner

You installed tawk.to because it was free. Genuinely free. Unlimited agents, unlimited chats, unlimited websites, no credit card, no trial countdown. In a market where everyone else wants 29 dollars a seat before you have said hello to a single customer, that is a legitimately great offer and I am not going to pretend otherwise.

Then a client looked at your site and said "what is tawk.to?"

And there it is. A little badge sitting under your chat window advertising somebody else on your own website. Which is fine, honestly. That is the deal, and it is a fair deal. Free software, small logo. Everybody understands the arrangement.

Until you go and check what it costs to remove it.

The Number

Removing tawk.to branding costs 39 dollars a month if you pay monthly, or 348 dollars a year if you pay annually. That annual figure works out around 29 dollars a month.

Sit with that for a second. Three hundred and forty eight dollars a year. To delete a logo.

Now, before anyone accuses me of being unfair, that is genuinely how freemium works and tawk.to are not hiding it. It is published on their help centre. Nobody is being tricked. If a free product wants to charge for white labelling, that is a completely normal business model and plenty of companies do exactly the same thing.

But here is the part that catches people out, and it is the reason I bothered writing this.

It Is Charged Per Property

The branding removal is not an account upgrade. It is a per property add on.

Property means website. So if you run one site, you pay 348 a year. If you run three sites, you pay it three times. Freelancers and agencies, I want you to do this maths slowly because it stings.

Websites	Annual cost to remove branding
1 site	348 dollars
3 sites	1,044 dollars
5 sites	1,740 dollars
10 sites	3,480 dollars
18 sites	6,264 dollars

That last row is not a made up number. We know an agency running eighteen client sites. If they wanted every client widget to look like it belonged to the client, that is six thousand two hundred dollars a year on a product they picked because it was free.

And the video, voice and screen sharing add on is another 49 a month or 348 a year. Also per property. You can see how this stacks.

The Costs That Are Not On The Pricing Page

Money is the easy part to measure. There are two other things worth knowing about before you commit.

The widget is enormous

DebugBear tested 21 chat widgets properly, in isolation, measuring what each one costs a page. Tawk.to came in at 749 kilobytes of JavaScript. That was the heaviest of every widget they tested. Heavier than Zendesk. Heavier than Intercom. Heavier than everything.

There is a thread on tawk.to's own community forum titled something like "chat widget causing slow page load", which tells you people have noticed.

Free in money. Three quarters of a megabyte in bandwidth, on every page view, for every visitor, including the ones who never open the chat.

The notifications

This is the complaint that shows up most often across G2, Capterra and the various review sites. Delayed notifications on the mobile app. Chats that arrive late or do not surface at all. People finding conversations hours after the visitor gave up and left.

Which is the thing that actually matters, right? A chat tool that does not tell you someone is chatting is not really a chat tool. It is a form that collects disappointment.

And it is a bit ironic, because tawk.to's own answer to the coverage problem is to sell you human beings. Their hired agent service is around one dollar an hour for 24/7 cover, which sounds cheap until you multiply it out to roughly 728 dollars a month. That is their business model working exactly as designed. The free product creates a gap, and the paid service fills it.

What You Are Actually Buying With That 348

Let me be fair here, because a hit piece is boring and tawk.to has real strengths.

What you get for free is genuinely unmatched. Unlimited agents is not a thing anywhere else. Unlimited chats with no conversation counter ticking down is not a thing anywhere else. Tidio caps its free plan at 50 conversations a month. Crisp gives you two seats. Tawk gives you everything and asks for nothing.

They have something like 166,000 companies using them, roughly 20 percent of the entire live chat market. You do not get there by scamming people.

So the question is not "is tawk.to bad." It is not bad. The question is narrower than that. It is: once you decide you want your own branding, is 348 dollars a year per site still the best deal available?

And that is where the maths gets interesting, because the moment you start paying, you are no longer comparing against free. You are comparing against every other paid option, and 348 a year buys quite a lot elsewhere.

What 348 A Year Buys Elsewhere

Roughly 29 dollars a month is a real budget in this market. Here is what sits around that price:

• Crisp Mini is 45 a month for 4 seats, so more expensive, but you get a proper CRM and audio calls.
• LiveChat Starter is 19 a month but only allows one user, which is useless the moment you hire.
• Chatwoot Startups is 19 per agent per month, which gets expensive fast with a team.
• Our own Pro plan is 12 a month with white labelling included, unlimited seats, and an AI agent. Agency is 29 a month and covers ten client workspaces, which is the row in that table above that hurts the most.

I have an obvious bias there and you should discount my opinion accordingly. But the structural point holds regardless of which tool you pick: paying per website for branding removal is an unusual pricing model, and it punishes exactly the people who use the product most.

A Word On The Removal Hacks

If you search for this, you will find CSS tricks and JavaScript snippets that hide the badge without paying. Some of them still work.

Do not do it. Two reasons and neither is moral lecturing.

First, it is against their terms, and accounts do get suspended. Losing your entire chat history to save 29 a month is a bad trade.

Second, and more practically, these hacks break. The vendor ships an update, the class name changes, and the badge quietly reappears on your client's website on a Tuesday while you are on holiday. You will not notice. Your client will.

If the branding bothers you enough to hide it, it bothers you enough to solve properly.

How To Decide In About Two Minutes

Answer three questions honestly.

One. How many websites? If it is one, and the badge does not bother you, stay on tawk.to. Seriously. It is free and it works and you have better things to worry about. If it is three or more and you want clean branding, the per property model is going to hurt and you should price alternatives.

Two. Do you actually see the messages? Go check right now whether there are unanswered chats sitting in your dashboard from last week. If there are, no pricing decision matters until you fix notifications, because the tool is not doing its job. Whatever you switch to, the test is whether a message reliably reaches a human within a couple of minutes.

Three. Does page speed matter to you? If you are running a content site that lives on search traffic, 749 kilobytes is a real cost. If you are running an internal tool or a site nobody finds on Google, it genuinely does not matter and you can ignore that whole section.

The Multi Site Workaround, And Why It Is Not Great

People running several sites sometimes ask whether they can just use one property across all of them and dodge the per site charge.

Technically you often can. Point the same widget at multiple domains and you are paying for branding removal once. Plenty of people do exactly this.

The problem is that it makes a mess of everything downstream. All your conversations arrive in one bucket with no reliable way to tell which site they came from, so you are guessing context on every message. Reporting becomes useless because you cannot separate one client from another. And if you are an agency, you now have all your clients' customer conversations mixed together in a single account, which is an awkward thing to explain if anyone asks about data separation.

It also breaks the moment a client wants their own access. You cannot hand over one site's conversations when they are tangled with four other businesses.

So the workaround exists, it saves money, and I would not use it for client work. For three side projects of your own, fine, go ahead.

The Honest Summary

Tawk.to is a good free product with an expensive exit ramp.

The free tier is more generous than anything else out there and I would recommend it to somebody who wants chat on one site tomorrow with no budget. That is a real recommendation, not a setup for a pitch.

But the second you have more than one website, or you need branding to look professional, the pricing model turns against you in a way that most people do not see coming until the invoice arrives. Per property add ons scale linearly with your success, which is the opposite of what you want from a tool you picked because it was free.

What Everyone Else Gives You For Free

Since the whole appeal here is the free tier, it is worth seeing how it stacks up. Free plans across the market, as they stand right now:

Tool	Free plan limits	Branding removal
Tawk.to	Unlimited agents, chats and sites	348 a year per site
Tidio	50 conversations a month, 10 seats	Around 17 a month on Growth
Crisp	2 seats, 100 customer profiles, no AI	Requires the 295 a month Plus plan
Chatwoot	2 agents, 500 conversations, 30 day history	Paid plans only
Freshchat	Up to 10 agents	Paid plans only
TGLiveChat	100 chats a month, 1 widget	Included from 12 a month

Look at that table honestly and tawk.to's free tier is the most generous by a distance. Unlimited anything is rare and they genuinely deliver it. Crisp giving you two seats and no AI credits is far more restrictive. Tidio's 50 conversations disappear faster than you would think, partly because automated greetings count toward the total.

So the free tier argument is won, comfortably. The paid comparison is the one that goes the other way, because everyone else charges per account and tawk charges per property.

Ask About Your Data Before You Commit

One more thing that is easy to overlook while you are comparing prices.

Whatever tool you use, at some point you may want to leave, and the ease of that varies wildly. Crisp's free tier famously does not let you export transcripts at all. Intercom's interface export gives you metadata without the actual message content, so you need the API for anything useful. Zendesk requires you to contact support to switch exports on before you can use them.

Before you pick anything, find the export button. Click it. See what comes out. If the answer is "there is no export" or "contact us", factor that in, because a tool you cannot leave is a tool that can raise its price whenever it likes.

This is not a hypothetical worry. Tidio repriced existing customers in December 2024 and people reported bills doubling with cards charged automatically. JivoChat users have reported sudden large increases. The time to check your exit is while you are calm and choosing, not while you are annoyed and leaving.

Work out your number using the table above. If it is under 350 a year, fine, pay it and move on. If it is over a thousand, you are now firmly in paid software territory, and you should shop like it.