Picking free live chat in 2026 is harder than dropping in tawk.to suggests

I compared ten products that all put a chat bubble on your site but sit in your stack very differently: tawk.to, Crisp, Freshchat, HubSpot, Tidio, Smartsupp, JivoChat, LiveChat, Zendesk, and Intercom. The quick forum answer is still “just use tawk.to” or “HubSpot if you already have CRM.” That skips the part that actually matters: agent caps, conversation limits, vendor branding, whether WhatsApp lives on the free tier, and what happens when you need EU hosting or SSO six months later.

For a solo operator or side project that needs to stay at zero cost, my default is tawk.to. Official materials advertise unlimited agents, unlimited chat volume, unlimited sites, unlimited history, and no concurrent-chat ceiling on the core product. You pay for branding removal, voice/video/screensharing, and heavier AI usage, not for basic live chat.

If EU-only data hosting and a cleaner shared-inbox path matter more than maximizing free breadth, I look at Crisp instead. The free tier caps you at two seats, but workspace pricing and a strong GDPR posture make it a serious small-team option once you outgrow solo work.

For a CRM-first tiny team already living in HubSpot, HubSpot free chat can work, but only with eyes open: two users, one inbox, 1,000 contacts, simple bots, and HubSpot branding on the widget. For a real support team in the 5-20 agent range, Freshchat is my balanced default because the free plan covers up to 10 agents before Growth tiers kick in at $19/agent/month billed annually.

At 50+ agents, I stop optimizing for free tiers. Zendesk is the steadier support-operations default. Intercom is the better fit for product-led SaaS that wants messaging, workflows, and AI woven through the customer journey, if you can model seat fees plus per-Fin-outcome billing.

Seven vendors here have a permanent free tier: tawk.to, Freshchat, HubSpot, Crisp, Tidio, JivoChat, and Smartsupp. LiveChat, Zendesk, and Intercom offer 14-day trials only. I did not find a permanent free Zendesk Chat plan in current official materials.

I prioritized official docs, pricing pages, and trust centers from mid-June 2026. Where docs were thin, I say so instead of guessing.

You cannot compare these on “free” alone

“Free” in this category spans several distinct economic models, and the headline number rarely tells you when you will hit a paywall.

flowchart LR
    need[Need website chat] --> model{Pricing model}

    model --> addon[Free core plus add-ons]
    addon --> tawk[tawk.to]

    model --> workspace[Flat workspace]
    workspace --> crisp[Crisp]

    model --> perAgent[Per-agent freemium]
    perAgent --> fresh[Freshchat]
    perAgent --> jivo[JivoChat]
    perAgent --> livechat[LiveChat]

    model --> volume[Volume-gated freemium]
    volume --> tidio[Tidio]

    model --> crmBundle[CRM bundle]
    crmBundle --> hubspot[HubSpot]

    model --> trial[Trial into paid suite]
    trial --> zendesk[Zendesk]
    trial --> intercom[Intercom]

Model	Vendors	What you trade
Free core plus add-ons	tawk.to	Unlimited core chat vs branding, voice/video, AI add-ons
Flat workspace	Crisp	Predictable team pricing vs tighter free seat cap
Per-agent freemium	Freshchat, JivoChat, LiveChat	Cheap at tiny scale vs compounding seat cost
Volume-gated freemium	Tidio	Generous agent count vs conversation and flow ceilings
CRM bundle	HubSpot	Free chat tied to CRM limits and HubSpot identity
Trial into suite	Zendesk, Intercom, LiveChat	Full product trial vs no forever-free path

The questions I ask before picking one:

Question	Why it matters
Forever-free or realistic pilot?	HubSpot and tawk.to stay free; Zendesk and Intercom do not
Agent count in 12 months	Freshchat free covers 10; Crisp free covers 2; JivoChat free covers 5 concurrent
Omnichannel now or later?	WhatsApp and social often start one tier above free
CRM-first, support-first, or product-led?	HubSpot, Freshchat/Zendesk, Intercom respectively
EU hosting, SSO, or SCIM?	Crisp and Smartsupp are EU-only; SSO is rarely on free tiers
AI billing model	Intercom charges per Fin outcome; Zendesk ties AI to resolution usage

Free does not always mean suitable for a real team. HubSpot is free forever, but two users and one inbox get tight fast. Expensive does not always mean bad value either. Intercom and Zendesk make sense when governance, help-center maturity, or regional hosting matter more than raw chat volume cost.

Free tiers at a glance

Provider	Permanent free	Key free limits	Paid entry	Best fit
tawk.to	Yes	Unlimited agents, volume, sites, history, concurrent chats; tawk.to branding unless removed	Core stays free; Remove Branding $39/mo or $348/yr; Video + Voice + Screensharing $49/mo or $348/yr	Zero-budget support, agencies, micro-SaaS
Freshchat	Yes	Up to 10 agents; 100 campaign contacts/account/month; WhatsApp and social on Growth	Growth $19/agent/mo annual; Pro $49; Enterprise $79	Small teams piloting support, then moving to omnichannel
Crisp	Yes	Up to 2 seats; unlimited conversations; no Hugo AI or advanced automation; free history retention unspecified	Mini ~$45/workspace/mo; higher tiers $95 and $295	Solo operators wanting EU hosting and flat workspace pricing
HubSpot	Yes	2 users, 1 inbox, 1,000 contacts; simple bots only; 3 active chats per user default	Starter from $20/seat; promo $10/seat observed on pricing page	CRM-first freelancers already on HubSpot
Tidio	Yes	Up to 10 agents; 50 conversations/month; flows for 100 unique visitors/month; 1 Lyro Guidance and 1 Smart Action; history retention unspecified	Paid plans from ~$29/mo depending on chat, flows, and AI mix	Ecommerce shops with moderate volume and automation needs
JivoChat	Yes	Up to 5 agents online at once; fewer features on Free; chat export on Professional and Enterprise only; branding/history limits unspecified	Basic $28/agent/mo; Professional $42; Enterprise $56	Messenger-heavy SMBs that also want telephony or WhatsApp
Smartsupp	Yes	Free tier exists; agent caps, history retention, and message caps unspecified in reviewed docs	Solo $17/mo annual; tiers at $25, $83, $280	Ecommerce teams wanting session replay and EU-only storage
LiveChat	No	14-day trial; Starter has 1 user and 60-day history	Starter $19/agent/mo annual; Team $49; Business $79	Teams wanting a specialist paid live-chat product
Zendesk	No permanent free tier found	14-day trial; live chat inside Suite with ticketing, help center, voice, AI	Plans from $19/mo; exact chat-equivalent cost depends on Suite package	Maturing support ops with governance and regional hosting needs
Intercom	No	14-day trial; seat pricing plus usage-based Fin outcomes; regional hosting on Advanced/Expert annual only	Essential $29/seat/mo + $0.99 per Fin outcome; Advanced $85; Expert $132	Product-led SaaS wanting messaging, automation, and AI in one stack

tawk.to has the broadest genuinely free support surface in this set: live chat, ticketing, knowledge base, contacts, APIs, and mobile apps, all without agent or volume caps on the core product. Data at rest sits in Google data centers in the United States. The tradeoff is vendor branding and US hosting unless you pay for removal or accept the data posture.

Freshchat is the most generous mainstream team-oriented free plan I found. Ten agents on free is enough to run a real pilot. WhatsApp, Facebook Messenger, Instagram, and richer routing wait on Growth. Freddy AI pooled usage beyond paid entitlements is not a lasting free perk.

Tidio looks generous on paper with 10 agents, but 50 conversations per month and 100 flow visitors per month force an upgrade faster than Freshchat for any site with real traffic. Lyro and flows are the upsell engine once those ceilings bite.

Crisp free is tighter on seats than tawk.to or Freshchat, but unlimited conversations on two seats can still work for a solo operator or pair. Unlimited chat history starts on Mini, so treat free history retention as unspecified until you verify in-product.

Smartsupp and JivoChat both have free tiers, but the reviewed official snippets did not clearly expose agent caps, history limits, or branding rules. Model costs conservatively and confirm inside the product before you plan around them.

Pricing models and what forces an upgrade

The procurement mistake I see most often is comparing headline prices without understanding the pricing logic. These vendors fall into five distinct models, and the model determines when you actually pay.

tawk.to is free core plus paid add-ons. Team size and chat volume are not the trigger. Branding control, voice/video/screensharing, and heavier AI Assist usage are. Remove Branding runs $39/month or $348 annually. Video + Voice + Screensharing is $49/month or $348 annually. AI Assist has a free Hobby tier with usage caps and paid tiers above that.

Crisp is flat workspace pricing. The upgrade path usually follows operational maturity: more than two seats, private notes, proactive triggers, shared email, custom domain, unlimited history, then ticketing and advanced analytics. Mini starts around $45/workspace/month on the reviewed pricing page.

Freshchat, LiveChat, and JivoChat bill per agent. At 2-5 agents with modest volume, that can look economical. At larger team sizes with stable workload, Crisp’s workspace model can beat per-agent compounding.

Tidio is volume-gated freemium. Fifty conversations per month or 100 flow visitors per month is the obvious forcing function. Paid packaging from about $29/month varies by how much live chat, flows, and AI you need.

Intercom blends seat fees with usage-based AI outcomes. Essential is $29/seat/month plus $0.99 per Fin outcome on the reviewed pricing page. Your AI success can become a line item on the invoice. Zendesk is moving in a similar direction with AI-resolution economics layered into Suite packaging.

HubSpot breakpoints are more users, more inboxes, branding removal, and bot logic beyond simple ticket/lead/meeting flows. Starter shows $20/seat in the product catalog, but I also saw a temporary $10/seat promotional price on the public pricing page at research time. Recheck before you buy.

A separate caution on “free” omnichannel claims: the cheapest entries often cover the web widget and maybe email, while WhatsApp, Messenger, richer automation, private help centers, SLA policy, custom routing, and identity controls sit one or two tiers higher. That pattern is especially visible in Freshchat, HubSpot, Intercom, and Zendesk, and partially true in Crisp as well.

Agent-based tools are often cheaper at very small scale. Workspace pricing can win once team size grows and per-agent math starts to hurt. Intercom deserves spreadsheet modeling before you commit because seat fees and per-outcome AI billing compound in ways a simple per-agent comparison will miss.

Features, integrations, and what actually differentiates

The useful split is whether the product is just a website widget or a broader customer-operations layer.

tawk.to is the broadest free support surface: chat, tickets, hosted knowledge base, contacts, unlimited history, JavaScript API, REST API, and mobile apps. I did not find a public mobile SDK in the reviewed docs. The guidance points developers toward embedding the widget or a webview instead.

HubSpot turns chat into a front door to CRM, inbox, and ticketing. The developer platform, APIs, SDKs, and marketplace are unusually strong for a product with a free entry point. Setup is heavier than “paste a snippet” once you start routing, bot logic, and identity assumptions.

Freshchat is support-desk-centric with a unified agent workspace and a clearer omnichannel roadmap on paid tiers. Freddy AI lands on paid plans. APIs and SDKs are mature enough for real integrations.

Crisp starts as a shared inbox and scales toward automation, ticketing, analytics, and multichannel ops on higher tiers. Hugo AI and advanced workflows sit above Free. The integration catalog is broad for SMB tooling: WordPress, Shopify, WooCommerce, Webflow, WhatsApp, Instagram, Slack, Salesforce, HubSpot, Zoho, Zapier, n8n, Twilio SMS, and more.

On automation and AI, the strongest official positioning in the reviewed materials came from Intercom, Zendesk, Freshchat, Crisp, HubSpot, and Tidio. Intercom puts Fin at the center with seat pricing plus per-outcome economics. Zendesk bundles AI agents into Suite with pricing tied to successful automated resolutions. HubSpot limits free bots to ticket, lead, and meeting use cases without custom branching. Tidio treats Lyro and flows as a major upsell once free thresholds are exceeded.

For multichannel messaging, Freshchat, Intercom, Zendesk, Crisp, and JivoChat stood out. Freshchat places WhatsApp, Messenger, Instagram, and other digital channels beyond Free. Zendesk positions live chat inside a broader Suite that includes voice. JivoChat collapses live chat, social messengers, email, and calls into one operator environment, which matters if your widget is really a sales and telephony hub.

For developer extensibility, HubSpot, Intercom, LiveChat, Crisp, and Freshchat publish the maturest API and SDK surfaces in this set, including web and mobile integrations. tawk.to’s JavaScript and REST APIs are meaningful but narrower on the mobile side.

A few feature differentiators matter more than marketing copy suggests. Crisp Magic Browse is one of the clearest native co-browsing-style features here, with explicit documentation on consent, privacy defaults, and the fact that it does not expose the visitor’s desktop or files. Smartsupp session video recordings are unusually useful for ecommerce troubleshooting and conversion analysis. tawk.to sells video, voice, and screensharing as an add-on. JivoChat offers telephony and separate video-call modules.

On analytics, tracking, mobile apps, and file transfer, nearly every serious vendor supports some mix of operator apps, visitor tracking, and conversation analytics, but documentation clarity varies. HubSpot documents website activity, dashboards, and chat-capacity controls. Crisp documents event tracking via its SDK and customer ratings. LiveChat and JivoChat both document file transfer with a 10 MB limit in agent apps. tawk.to documents Geo-IP, engagement tracking, and reporting.

Privacy, security, setup, and reliability

For buyers with real compliance requirements, data residency can matter more than price.

Provider	Hosting posture (from reviewed docs)
Crisp	EU-only; messaging data in the Netherlands, plugin data in Germany
Smartsupp	EU-only data storage
tawk.to	US Google data centers
HubSpot	US, EU, Canada, Australia; paid accounts can migrate to preferred region
Freshworks (Freshchat)	US, Europe, India, Australia; AWS-backed regional hosting
LiveChat	US or EU data-center choice
Zendesk	AWS regions across US, EEA, UK, Asia-Pacific; regional hosting tied to policy and entitlement
Intercom	US, EU, Australia on qualifying plans; region change requires a new workspace
JivoChat	US, Southeast Asia, EU
Tidio	Encryption and GDPR/CCPA alignment documented; specific data region less explicit

On core security controls, the broad direction in vendor docs is reassuring, but tier placement varies. tawk.to states encryption in transit and at rest with retention until deletion. Freshworks documents AES-256 at rest and TLS 1.2 in transit. Tidio documents HTTPS/TLS, encryption at rest, hashed credentials, and GDPR/CCPA alignment. LiveChat documents 256-bit SSL, secure storage, and credit-card masking. HubSpot and Zendesk frame security around trust-center materials rather than simple feature lists. Intercom’s security docs surface SSO, SCIM, 2FA, IP restrictions, and formal reliability language most explicitly in this set.

SSO is rarely a free-tier feature. LiveChat, Freshworks, HubSpot, and Intercom publish the clearest SSO documentation. Zendesk’s trust materials position it as an enterprise governance platform, though exact tiering is packaging-dependent. Crisp documents 2FA clearly; I did not find an equally clear public SSO doc during this pass. For tawk.to, Tidio, Smartsupp, and JivoChat, SSO was not prominent in the reviewed free-first materials. Validate federated identity requirements before you commit.

Setup complexity varies less than buyers assume on day one and more than they expect by week two. The initial widget install is easy almost everywhere: copy-paste JavaScript or a CMS plugin, and Tidio markets roughly five-minute setup. Complexity shows up in routing logic, channel governance, identity, analytics, data model, and regional hosting. HubSpot is moderately heavier than its “free chat” branding suggests. Zendesk and Intercom are the heaviest to operationalize well.

On performance and reliability, leading vendors all document serious infrastructure, but only some make reliability a visible buying argument. tawk.to highlights globally distributed architecture on Google Cloud. Crisp runs EU-hosted core storage and relay servers outside the EU for latency. Intercom Fin claims redundancy across regions and a 99.8% SLA in reviewed materials. Free tools are not automatically unreliable, but formal operational commitments and governance get much clearer on higher-end products and tiers.

A 2024 TechRadar review of tawk.to called it exceptional value for a free live-chat solution while noting that some competitors are better on mobile and AI depth. That matches what I saw in primary sources: tawk.to wins on free-core value, but the strongest AI, governance, and multichannel depth sit elsewhere.

What to pick by team size

Solo freelancer or side project

My default is tawk.to. No other reviewed vendor makes a stronger official claim around zero-cost endurance for a single operator or tiny team. Unlimited agents, history, volume, sites, and concurrent chats remove the classic fear that “free” is really a short pilot.

Choose Crisp instead if you are privacy-sensitive, EU-centric, or you expect to graduate from solo work into a structured shared inbox and prefer flat workspace pricing over maximizing free breadth.

Choose HubSpot only if website chat is really a lead-capture surface for an already-HubSpot-centric motion. Two users and one inbox are fine for a freelancer; they are not fine for a growing support team.

Smartsupp is worth a look for ecommerce side projects where paired session video recordings help you debug checkout friction, assuming EU-only storage fits your posture and you verify the unspecified free-tier limits in-product.

Small business with 5-20 agents

My default is Freshchat. The free plan lets you pilot with up to 10 agents. Growth at $19/agent/month unlocks more channels, dashboards, routing, and administration without forcing an enterprise deployment on day one.

Tidio is the best lower-end challenger if your support motion is ecommerce-heavy and real conversation volume stays modest. The 50-conversation and 100-flow-visitor free ceilings mean you usually discover the paid plan faster than with Freshchat.

Crisp deserves a serious cost model at this team size if you want flatter pricing and clear EU data posture, especially once per-agent tools start compounding.

JivoChat fits sales-led or messenger-heavy SMBs that also want calling or WhatsApp in the same operator environment. Free covers five agents online at once, but export flexibility is weaker until Professional or Enterprise.

Startup scaling to 50+ agents

My default is Zendesk. It is the steadier choice when your future state looks like a real support operation: ticketing, live chat, help center, voice, AI, security review, and region-aware hosting.

Intercom is the better choice when the support widget is part of a broader customer-journey system spanning onboarding, product support, proactive engagement, and AI handoff. Model Fin usage carefully. Seat fees plus per-outcome billing can produce a better customer experience and a more complicated invoice at the same time.

LiveChat remains excellent if you specifically want best-in-class paid live chat, but it is not my first recommendation as the long-term system of record for a 50+ agent support org.

Migration and decision checklists

A migration goes smoother when you treat the widget as the last step, not the first. Confirm your data model, exports, routing, identity assumptions, and success metrics before you swap the snippet.

Crisp markets migration of chats, help-center articles, and contacts. Tidio publishes migration guides from Zendesk, LiveChat, and Intercom. Intercom’s regional-hosting documentation says changing region requires a new workspace and manual transfer, not an in-place migration. JivoChat chat export is limited to Professional and Enterprise, which can materially affect exit planning.

flowchart LR
    A[Audit channels, agents, KPIs, privacy requirements] --> B[Export chats, contacts, articles, macros]
    B --> C[Clean data and retire obsolete rules]
    C --> D[Map inboxes, tags, automations, and ownership]
    D --> E[Pilot on one site or one channel]
    E --> F[Train agents and run dual support briefly]
    F --> G[Cut over widget, snippets, channel routing]
    G --> H[Validate history, analytics, consent, and escalation]

Migration checklist:

Confirm what must move: chats, contacts, tickets, articles, macros, tags, CSAT history, and analytics baselines.
Confirm what cannot move cleanly: social-channel history, region moves, AI training data, and bot logic are the most common problem areas.
Freeze your taxonomy before export. Teams that migrate bad tags and bad canned replies only recreate the old mess.
Pilot with one web property or one channel first.
Run agent training on routing, handoff rules, mobile usage, and privacy prompts before cutover.
Preserve your old KPI baseline for at least one reporting cycle so you can tell whether the new tool is genuinely better.

Decision checklist:

Do you need forever-free, or just a realistic pilot?
How many agents will you have in 12 months, not just at launch?
Do you need WhatsApp, Instagram, Facebook Messenger, SMS, or voice now, or only later?
Is your center of gravity CRM-first, support-first, or product-led messaging?
Are EU-only hosting, regional residency, SSO, or SCIM mandatory?
Do you need co-browsing, session replay, or voice/video escalation?
Are you comfortable with usage-based AI pricing, or do you need cost predictability?

Gaps in the public docs

Smartsupp and JivoChat were the thinnest on free-tier details. I could not find clear public numbers for retention, branding removal, or concurrent chat limits on either site. Crisp and Tidio were better, but still left gaps on the items that usually decide whether “free” survives past launch week.

HubSpot had promo pricing on the public page when I wrote this. Treat any dollar figure as a snapshot, not a quote.

I did not run a side-by-side delivery test across all ten vendors. Reliability notes here come from each vendor’s docs and trust pages. TechRadar’s 2024 tawk.to review was the only third-party write-up I leaned on heavily.