BIK
Also known as: Bik.ai, BIK Global
Agentic AI CRM for eCommerce that consolidates marketing and support: 100+ prebuilt AI commerce agents plus an Agent Studio to build your own, deployed across WhatsApp, Instagram, web, Messenger, SMS and email with a deep Shopify integration and a public REST API.
BIK is an agentic AI CRM for eCommerce, built by the team behind Manifest AI, that consolidates a brand's marketing and support stack under one dashboard so merchants can deploy AI commerce agents rather than stitching together plugins or adding headcount. It ships more than 100 pre-built AI commerce agents covering acquisition, retention, order management, helpdesk and personalization, alongside an Agent Studio where a brand builds its own by giving an agent a goal, success criteria and instructions. BIK says more than 500 global brands run on the platform, driving over $100m in attributed revenue.
The platform is organized as a set of capability modules: agentic acquisition and retention, AI order management, an AI helpdesk, agentic gamification, an agentic customer data platform, advanced analytics, AI campaigns, and Agent Studio, with Manifest AI available alongside for on-site shopping assistance. Agents deploy across WhatsApp, Instagram, a website chatbot, Facebook Messenger, SMS and email, and the integration with Shopify runs deep, tying two-way messaging to store events on revenue flows such as abandoned cart recovery, upsell, cross-sell and welcome, plus Instagram comment-to-DM, AI product recommendations drawn from the live catalog, and gamification like spin the wheel, back-in-stock and price-drop alerts.
The support side unifies an inbox across Messenger, Instagram, WhatsApp, SMS and email where human and AI agents work together, with AI Reply drafting responses, AI Summary condensing threads, AI Translate handling foreign-language tickets in real time, and AI Classify routing on intent. Campaign delivery is tuned by spam protection, rate limiters and reattempt logic.
For a technical team the platform is more open than its marketing suggests. BIK publishes public developer documentation with REST APIs for messaging, customer profiles, product catalog sync, event ingestion and flow triggering, plus webhooks and a Postman collection, so agents and journeys can be driven from a brand's own systems. Security is claimed rather than evidenced: SOC 2, GDPR and ISO marks appear on the homepage with no report, standard number or audit detail published, and no SSO, role-based access control or audit logging is documented.
For a Shopify brand that wants marketing and support agents on one platform with a real API underneath, BIK is a strong and unusually consolidated fit. A buyer with formal security review requirements, or one who needs to choose the underlying model, test an agent before it goes live, or know where their data is hosted, will find those answers are not published.
Vendor details
Canonical URL
https://bik.ai/
Category
Customer support agent
Funding status
Y Combinator backed (Summer 2020 batch), founded 2019 by Sonakshi Nathani and Ashutosh Singla, and built alongside sibling product Manifest AI. BIK states on its own homepage that more than 500 global brands use the platform, driving over $100m in attributed revenue. Specific funding round sizes are not disclosed on any first-party source.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Integrates natively with Shopify for two-way conversational messaging tied to store events, AI product recommendations drawn from the connected catalog, and order actions through an AI Order Management module. Beyond Shopify, BIK describes itself as an open API platform: public developer documentation at docs.bik.ai publishes REST endpoints for messages across WhatsApp, email and SMS, customer profile creation and update, product catalog sync, event ingestion for cohort building, and flow triggering, plus webhooks for real-time notifications and a Postman collection. Channels span WhatsApp, Instagram, a website chatbot, Facebook Messenger, SMS and email.
Sources & related URLs
Agentic Index coverage score
8.0 / 14 capabilities · 57%
| Integrations & Tool Calling | Full |
|---|---|
|
Named systems plus authenticated write action, on the vendor's own surfaces. BIK integrates natively with Shopify for two-way conversational messaging tied to store events, AI product recommendations drawn from the connected catalog, and order actions through an AI Order Management module. Beyond Shopify the homepage states that any app can be connected, describing BIK as an open API platform using webhooks and REST API blocks, and the developer docs make that concrete with documented endpoints to create and update customer profiles, sync the product catalog, ingest events and trigger flows, plus webhooks for real-time notification back out. A dedicated integrations page is published at bik.ai/integrations. Sourcedocs.bik.airead 2026-09-05 |
|
| Workflow Orchestration | Full |
|
Multi-step journeys with a documented trigger interface. BIK publishes journey and campaign builders covering revenue flows including abandoned cart recovery, order confirmation, welcome, upsell and cross-sell, and the developer docs expose a Flows API to trigger automated journeys programmatically alongside an Events API that ingests customer events to build the cohorts those journeys act on. Agent Studio adds the agent-level construct: a brand defines an agent's goal, success criteria and instructions, which is delegated objective rather than a fixed script. Work moves between surfaces, with AI Classify routing on intent and handoff into the shared inbox when a person is needed, and a drip module sequences follow-up over time. Sourcedocs.bik.airead 2026-09-05 |
|
| Knowledge Grounding & RAG | Partial |
|
Grounding is catalog-shaped and the corpus behind it is not described. Agents draw on the connected Shopify catalog, with AI Product Recommendation stated to work off that integration so suggestions match what the store actually sells, and the Products API keeps the catalog synced as a maintained source. Agents are configured with instructions and brand context so responses stay on-brand, and AI Classify interprets intent against that setup. No document knowledge base, help center ingestion, retrieval architecture, refresh behavior or source citation is documented, so the grounding is structured commerce data rather than a knowledge layer a support agent answers policy questions from. Sourcebik.airead 2026-09-05 |
|
| Human Oversight & Guardrails | Partial |
|
Humans and agents share a workspace, but no control surface over the agent is documented. BIK publishes a multi-channel CRM and AI Helpdesk where human and AI agents work the same inbox across WhatsApp, Instagram, Messenger, SMS and email, with AI Reply drafting responses, AI Assistant and AI Summary supporting the human, and AI Classify routing by intent so a case reaches the right resolution path. No guardrail configuration, content policy, approval step, autonomy ceiling or documented escalation rule is published. A shared inbox is where a person can intervene, not a mechanism that governs what the agent may do before they do. Spam limiters throttle outbound message volume, a campaign deliverability feature rather than oversight of an agent. Sourcebik.ai/multichannel-crmread 2026-09-05 |
|
| Security, Identity & Governance | Partial |
|
A homepage trust block, headed that customer data is BIK's first priority, displays SOC 2, GDPR and ISO marks alongside a Meta Business Partner badge, and the company publishes a dedicated GDPR page, a privacy policy, terms and a master agreement. The certifications are not evidenced beyond the badges: there is no trust portal, no report is downloadable or offered on request, no ISO standard number or SOC 2 type is stated, and no auditor is named. No customer facing controls are documented either, with no SSO, role based access control, audit log or retention configuration. Sourcebik.airead 2026-09-05 |
|
| Observability & Auditability | Partial |
|
Outcome reporting is documented; execution tracing is not. Advanced Analytics is published as one of BIK's named capability areas, covering campaign performance, real-time growth insight and revenue attribution, with the vendor's own headline claim resting on measurable ROI from those dashboards, and the Events API means the underlying activity stream is queryable by the brand. No per conversation reasoning trace, tool call log, session history view or configuration audit trail for an individual agent run is documented, so a buyer can see what an agent achieved in aggregate but not reconstruct what it did in a given case or who changed it. Sourcebik.airead 2026-09-05 |
|
| Memory & State Persistence | Partial |
|
A customer data platform is documented and an agent memory layer is not. BIK publishes an Agentic CDP module as one of its named capability areas, and the developer docs expose a Customers API to create and update profiles plus an Events API that ingests behavioral events to build cohorts and segments, so a persistent unified profile exists and follows the customer across WhatsApp, Instagram, SMS and email with conversation history attached. The CDP is a database the brand loads, segments and queries. No agent memory product is published, no scope, lifetime or deletion control for retained agent state is documented, and nothing describes what an agent itself carries between separate conversations. Sourcedocs.bik.airead 2026-09-05 |
|
| Deployment & Data Residency | Not documented |
|
The product pages state no hosting region, data center location, residency option or self hosted deployment. BIK's legal documents are a privacy policy at /bik-privacy-policy, a dedicated GDPR page at /gdpr, terms at /bik-terms and a master agreement at /master-agreement. Delivery is cloud only SaaS, installed either directly or through the Shopify App Store, with no self hosted or private cloud option marketed anywhere, and the vendor's repeated data ownership claim, that the brand owns its data, is a commercial position about rights rather than a statement about location. Sourcebik.ai/gdprread 2026-09-05 |
|
| Prebuilt Agents / Templates / Packs | Full |
|
A catalog the buyer browses, plus a builder for what it does not cover. The homepage states it twice: 100+ AI commerce agents across retention, acquisition and support, and choose from the 100+ pre-built AI commerce agents or build an agent of your choice by giving it a goal, success criteria and instructions. Agents are published across nine capability areas (agentic acquisition, agentic retention, AI order management, AI helpdesk, agentic gamification, agentic CDP, advanced analytics, Agent Studio and AI campaign), with prebuilt revenue flows and gamification modules alongside. The 500+ figure on the same homepage counts global brands using BIK, not agents. The homepage section meant to list the agents currently shows the empty state 'No workflows available at the moment', so the catalog is described there but not displayed. Sourcebik.airead 2026-09-05 |
|
| Triggers & Channel Coverage | Full |
|
Six channels and three documented ways for work to start. Channels: WhatsApp, Instagram, a website chatbot, Facebook Messenger, SMS and email, with the Messages API documenting WhatsApp, email and SMS delivery directly. Initiation is inbound through the helpdesk and chatbot; event-driven, with an Events API ingesting customer events and store activity that fire journeys, covering abandoned cart, back-in-stock and price-drop; and social, with Instagram comment-to-DM turning a public comment into a private conversation. Outbound campaigns and AI drip sequences give the agent an initiating path rather than only a receiving one, and the Flows API lets an external system start a journey on its own schedule. Sourcedocs.bik.airead 2026-09-05 |
|
| Model Flexibility & Routing | Not documented |
|
Agent configuration is stated as goal, success criteria and instructions, which is prompt level control rather than model level choice. No model selection, routing configuration or bring your own key path is published, the API reference at docs.bik.ai covers messaging, events, flows, customers and products with no model parameter, and no model provider is named. Sourcebik.airead 2026-09-05 |
|
| APIs / SDKs / MCP Extensibility | Full |
|
A public developer documentation site, linked from BIK's own footer. BIK publishes REST API documentation at docs.bik.ai with a getting-started guide, an authentication section, a full API reference and a Postman collection, all reachable without a login. Six named API groups are documented: Messages for WhatsApp, email and SMS delivery, Events for ingesting customer events to build cohorts, Flows for triggering automated journeys, Customers for creating and managing profiles, Products for syncing the catalog, and Webhooks for real-time event notifications. The homepage states the same position in plain terms, describing BIK as an open API platform connected through webhooks and REST API blocks. No client SDK or MCP server is published. Sourcedocs.bik.airead 2026-09-05 |
|
| Testing, Debugging & Optimization | Partial |
|
A target is set at build time and measured after launch; nothing tests the agent in between. Agent Studio has the customer define an agent's goal, success criteria and instructions, so a standard exists against which performance can be read, and Advanced Analytics reports outcomes once live. On the delivery side, spam protection, rate limiters and reattempt logic optimize whether a message lands, which is genuine optimization of the send. No simulation, staging environment, test case, regression suite or scored pre release evaluation of a proposed agent is documented, so nothing catches a bad agent before customers meet it. The deliverability modules act on message transmission rather than on agent behavior. Sourcebik.airead 2026-09-05 |
|
| Browser / Computer-use | Not documented |
|
The route out is programmatic: agents act through the native Shopify integration, through documented REST endpoints for messages, customers, products, events and flows, and through webhooks, and they reach people over conversational channels rather than by operating a screen. No browser control, hosted session, virtual desktop or interface automation is documented. Instagram comment-to-DM and in-chat purchase run on platform APIs, not on an agent driving the Instagram interface. Sourcebik.airead 2026-09-05 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Paid plans via the Shopify App Store, billed in USD; exact plan prices are not published in retrieved sources.
Monthly Shopify App Store subscription plans (USD), with likely per message channel fees
Cost watchouts
Conversational channels such as WhatsApp typically incur separate Meta conversation fees; some engagements have required advance payment per merchant reviews.
Variable cost rationale
Beyond the subscription, conversational channels such as WhatsApp typically carry Meta conversation fees that scale with message volume, though the exact pass through model was not confirmed in retrieved sources.
Sales call required
No, self serve available
Free / trial
Availability of a free tier or trial not confirmed in retrieved sources
Lowest paid plan
Not captured; see the Shopify App Store listing
Key ambiguities
Plans are listed on the Shopify App Store and billed in USD, but exact prices, free tier and trial availability were not captured in retrieved sources, and some engagements involve advance billing.
Related vendors
- Decagon — AI agent platform purpose-built for customer support, with deep…
- 5.Y — Singapore early-stage platform whose GLUCOSE product deploys…
- Ada — Agentic customer experience platform whose AI agents autonomously…
- Aisera — Enterprise agentic AI platform that orchestrates specialized agents…
- ASAPP — Generative AI for enterprise contact centers that plugs into…
- Assembled — AI customer support orchestration platform that unifies autonomous…
Alternatives to BIK
The closest documented capability profiles to BIK among customer support agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Spoki8.0 / 14Adds documented Deployment & Data ResidencyBIK vs Spoki →
- Readyly7.5 / 14Fuller documented coverage on Knowledge Grounding & RAG
- Siena AI7.5 / 14Fuller documented coverage on Memory & State Persistence
- DigitalGenius8.0 / 14Adds documented Deployment & Data Residency
- Gradient Labs7.0 / 14Fuller documented coverage on Human Oversight & Guardrails
- Maven AGI8.0 / 14Adds documented Deployment & Data Residency
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded