Frequently asked questions

Eshal works in three layers. First, your knowledge - documents, policies, FAQs, and connected systems - is ingested and indexed using RAG, so the AI always has accurate, current information to draw from. Second, your AI Concierge handles the conversation: understanding what a customer or employee is asking, maintaining context across multiple turns, and responding in the right language and tone. Third, AI Agents take action where needed - retrieving data, updating records, or completing a task - all within the guardrails and permissions your organization sets. The result is a single conversation that can both answer a question and resolve it.

Yes. Eshal is built for enterprise integration. Our platform connects with your CRM, helpdesk, ERP, knowledge management systems, and other enterprise tools - so your AI Concierge can access real-time data, retrieve customer records, and take meaningful actions, not just answer static questions. The questions below cover some of the specific platforms we connect to most often.

Yes. Eshal integrates with Microsoft 365, allowing your AI Concierge to work alongside the tools your teams already use - including connecting to SharePoint for knowledge and supporting Microsoft's identity tools for secure, single sign-on access. This is particularly valuable for internal-facing AI Concierges supporting HR, IT, or employee self-service.

Yes. Eshal integrates with Salesforce, giving your AI Concierge access to live customer and case data so it can answer account-specific questions, update records, and create or route cases directly from a conversation - keeping your CRM as the single source of truth while customers get instant, accurate answers.

Yes. Eshal integrates with SAP, enabling your AI Concierge to retrieve order status, inventory, and account information directly from your ERP and respond with live, accurate data rather than static answers. This is especially valuable for manufacturing, retail, and distribution organizations running SAP as their operational backbone.

Yes. Eshal integrates with Zendesk, Intercom, and other common helpdesk platforms, so your AI Concierge can create, update, and route tickets, and hand off conversations to a human agent with full context already in place - keeping your existing support workflow intact while adding an AI layer in front of it.

Eshal can be deployed across the channels your customers already use, including your website, mobile app, and common messaging platforms such as WhatsApp. Your AI Concierge maintains the same knowledge, persona, and guardrails across every channel, so customers get a consistent experience wherever they reach out. Contact our team to confirm specific channel availability for your deployment.

RAG stands for Retrieval-Augmented Generation. It is the technology that allows an AI to pull accurate, up-to-date information from your own knowledge base before generating a response. Eshal uses RAG so your AI Concierge answers questions based on your actual policies, product documentation, and internal content - not generic AI guesses. The result is highly accurate, brand-consistent responses your customers can trust.

Eshal is built to work with leading large language models, including GPT, Claude, Gemini, Llama, and Mistral. Rather than locking you into a single model, Eshal lets you choose the model that best fits your accuracy, cost, latency, and data residency requirements - and switch as your needs change, without rebuilding your AI Concierge from scratch.

Eshal grounds every response in your organization's approved knowledge using RAG, configurable guardrails, and governance controls - not generic internet information. Organizations can continuously review and update their knowledge base to improve accuracy over time. Eshal also surfaces unanswered or low-confidence questions so gaps can be identified and closed, meaning your AI Concierge gets more accurate the longer it is deployed.

Yes. Eshal answers using your organization's approved knowledge - your documents, policies, FAQs, and connected systems - rather than generic internet information. What your AI Concierge knows is determined entirely by what you configure and approve. Your data remains under your control.

Eshal reduces hallucinations through several layers working together. RAG grounds every response in your own approved knowledge rather than the model's general training data. Confidence thresholds mean that if Eshal isn't confident it has the right answer, it says so rather than guessing - and can escalate to a human instead. Configurable guardrails constrain what topics and types of response the AI Concierge is allowed to give. And every interaction is auditable, so low-confidence or unanswered questions are surfaced for your team to review, continuously tightening accuracy over time.

Training your AI Concierge means giving it the right knowledge to draw from, and Eshal makes this a no-code process. You can upload PDF documents, Word files, and FAQs, connect website pages and help centre articles, or sync directly from SharePoint, your CRM, and other databases. Once ingested, Eshal's RAG engine indexes the content and makes it retrievable, so your AI Concierge answers from accurate, current information every time - with no AI or technical expertise required to keep it up to date.

Eshal's AI Agents go beyond answering questions - they can complete tasks on behalf of your customers and teams. Depending on your configuration and integrations, this includes booking appointments, submitting and routing support tickets, retrieving live account or order information, updating customer records, processing requests, and triggering workflows in connected systems. Actions are governed by the permissions and rules you define, ensuring your AI Agent operates within the boundaries you set.

Yes. Beyond answering questions, Eshal can trigger and run multi-step workflows in your connected systems - for example, verifying a customer's identity, checking eligibility, updating a record, and confirming the outcome, all within a single conversation. Workflows are configured to match your existing processes, so Eshal extends what your systems already do rather than requiring you to rebuild them.

Yes. Eshal can capture visitor information, ask qualifying questions, and route qualified leads to your sales team or CRM automatically - turning routine website and chat traffic into a structured pipeline rather than requiring a human to triage every enquiry manually.

Voice is on Eshal's platform roadmap. Today, Eshal is primarily deployed across text-based channels - web, mobile, and messaging. If voice is a requirement for your use case, speak with our team about current capability and timeline.

Eshal supports seamless escalation to human agents when a conversation requires it. You define the conditions that trigger a handoff - such as query complexity, customer sentiment, specific topics, or explicit customer requests. When escalation occurs, the full conversation context is transferred to your agent so customers never have to repeat themselves. Handoff works across your connected helpdesk or live chat platforms, keeping the experience consistent and uninterrupted.

No. Eshal is designed to work alongside your team, not replace it. It handles the high volume of repetitive, time-consuming queries - freeing your human agents to focus on complex, high-value interactions that genuinely require human judgment and empathy. Most organizations find that Eshal significantly reduces support workload while allowing the same team to deliver a far better customer experience.

Yes. Eshal retains conversation history within a session, allowing the AI Concierge to maintain context throughout a conversation so customers never have to repeat themselves. When a handoff to a human agent occurs, the full conversation history is transferred alongside it. Data retention policies are configurable to meet your organization's compliance requirements.

Yes. Eshal supports multi-tenant deployments, making it suitable for enterprise groups, government bodies with multiple departments, or technology partners deploying Eshal across several clients. Each organization or business unit operates with its own knowledge base, configuration, and permissions.

Yes. Eshal supports running multiple AI Concierges within a single organization - for example, a dedicated HR Concierge, IT Concierge, Customer Service Concierge, and Sales Concierge, each with its own knowledge base, persona, and permissions, while still operating under one account and one set of governance controls. This is common among enterprise customers who want department-specific AI Concierges without managing separate platforms or vendors.

Yes. Eshal is built on cloud infrastructure designed to scale with demand, handling large volumes of concurrent conversations without a drop in response time or accuracy. Whether you're supporting a few hundred conversations a day or scaling to enterprise-level volume during peak periods, Eshal's architecture is built to handle the load.

Eshal is highly customizable. You can configure your AI Concierge's name, personality, tone of voice, and brand visual identity so it feels like a natural extension of your organization. Conversation flows, escalation rules, response styles, and topic boundaries are all configurable without requiring technical expertise. Enterprise customers can go further with advanced persona configuration and deep workflow customization.

Eshal gives you visibility into exactly how your AI Concierge is performing. You can track conversation volume, resolution rates, escalation rates, response accuracy, and customer satisfaction signals - all from a single dashboard. Eshal also surfaces knowledge gaps, the questions your AI Concierge couldn't confidently answer, so you know precisely where to improve. Enterprise customers have access to deeper reporting and custom dashboards.

Yes. Eshal is built to enterprise-grade security standards across every layer of the platform - from how data is encrypted in transit and at rest, to how access is controlled, to how your knowledge base is isolated from every other customer's. Our architecture is designed to meet the governance, auditability, and compliance requirements of regulated industries including healthcare, financial services, and government. The questions below cover the specifics; for a detailed security overview, contact us at security@eshal.ai.

No. Your data, conversations, and knowledge base are never used to train shared or public AI models. Eshal is designed with strict tenant isolation, meaning what one customer's AI Concierge knows and learns stays entirely within that customer's environment.

Eshal supports flexible data residency to meet the requirements of different organizations and jurisdictions. Data residency region can be configured at deployment, and for organizations with strict data sovereignty requirements - including government and regulated-industry customers - on-premise and private cloud deployment options keep your data entirely within your own infrastructure. Contact our team to discuss the right configuration for your organization.

Yes. In addition to standard cloud deployment, Eshal supports private cloud and on-premises, air-gapped deployment for organizations with strict data sovereignty, regulatory, or security requirements - including government bodies and regulated industries. The right model depends on your infrastructure and compliance needs; our team will help you determine the best fit during scoping.

Eshal's security architecture includes encryption in transit and at rest, role-based access control, tenant isolation between customers, detailed audit logging, and configurable guardrails that govern what your AI Concierge is allowed to say and do. Together, these controls are designed to give security and compliance teams the visibility and assurance they need before approving deployment.

Yes. Eshal is built around a three-tier guardrails architecture, meaning the boundaries on what your AI Concierge can discuss, what actions your AI Agents can take, and when a conversation must escalate to a human are all configured by your administrators - not left to the model's discretion. This is what we mean by compliance by architecture: governance is built into the platform itself, not applied as an afterthought.

Eshal's security controls are aligned with the compliance requirements enterprise and regulated-industry customers typically require, and we are actively pursuing formal certification, including SOC 2. For organizations with specific compliance requirements, our team can share detailed documentation of our current security posture and certification roadmap - contact us at security@eshal.ai.

Yes. Eshal is built with the governance, auditability, and data control that regulated industries require, and is already designed for use in healthcare, financial services, insurance, and government. See the Industries & use cases tab for how Eshal applies specifically to regulated sectors.

You do. Your data, your knowledge base, and everything your AI Concierge learns from your organization remain entirely your property. Eshal never shares your knowledge base with other customers, never uses it to train shared models, and provides full export capability so you are never locked into the platform.

Yes. Eshal supports a Bring Your Own Key (BYOK) model, allowing your organization to connect its own AI model provider account or API keys rather than relying solely on Eshal's. This gives you direct control over model selection, cost, and data handling - particularly relevant for financial services, government, and healthcare customers operating under strict procurement or data governance frameworks.

Yes. Eshal supports single sign-on integration with your organization's identity provider, allowing your team to manage access through existing enterprise credentials rather than separate logins. This simplifies onboarding, strengthens access control, and aligns with standard enterprise IT security requirements. Contact our team to confirm specific protocol support for your environment.

On termination, Eshal provides a defined data offboarding process. You retain the ability to export your data prior to termination, and your data is deleted from our systems within a specified period following contract end. Exact timelines are defined in your contractual agreement, and enterprise customers can negotiate specific data deletion and confirmation requirements as part of their contract.

Eshal provides data export capability so your organization is never locked in. You can export your knowledge base content, conversation history, and configuration data in standard formats. Enterprise customers can request structured data exports at any time during the contract period.

Eshal maintains an incident response process with defined notification timelines. In the event of a confirmed security incident affecting customer data, we notify affected customers without undue delay and in accordance with our contractual obligations and applicable data protection regulations. Enterprise and regulated-industry customers can negotiate specific incident notification timelines as part of their contract.

Government organizations use Eshal to make public services more accessible. Citizens get instant, accurate answers about services, eligibility, processes, and documentation requirements - 24 hours a day, in their preferred language - without waiting on call centres or navigating complex websites. Eshal reduces administrative burden on government teams while improving the citizen experience, and its enterprise-grade security architecture meets the stringent data governance requirements of public sector deployments.

Financial services firms use Eshal to deliver fast, accurate responses to customer queries about products, account information, processes, and policies - across digital and branch channels. Eshal reduces pressure on contact centres, improves first-contact resolution, and ensures every customer response is consistent, compliant, and drawn from approved organizational content. For firms navigating complex regulatory environments, Eshal's governed AI approach provides the control and auditability the industry demands.

Healthcare organizations use Eshal to handle high volumes of patient and visitor queries - appointment information, department directions, service availability, and policy questions - without adding pressure to clinical and administrative staff. Eshal delivers accurate, governed responses based on your own approved content, ensuring patients receive reliable information while your team focuses on care delivery. Our platform is built with the security and compliance requirements of regulated healthcare environments in mind.

Universities, schools, and training organizations use Eshal to handle the constant flow of queries from students, parents, and staff - admissions information, course details, scheduling, campus services, and administrative processes. Eshal reduces the load on administrative teams during peak periods like enrolment season, while ensuring prospective and current students always have access to accurate, up-to-date information. Multilingual support ensures accessibility across diverse student populations.

In hospitality, every guest interaction shapes the overall experience. Eshal enables hotels, resorts, and hospitality groups to deliver a personalized, always-available concierge experience across digital touchpoints - handling reservation queries, amenity information, local recommendations, check-in guidance, and service requests instantly. Staff are freed to focus on in-person guest experience while Eshal ensures no query goes unanswered, at any hour.

Retail organizations use Eshal to deliver fast, accurate customer support across digital channels - handling product questions, order status, returns, store information, and promotions without agent involvement. During peak trading periods, Eshal scales instantly to handle demand spikes that would overwhelm a human support team. The result is higher customer satisfaction, lower support costs, and agents freed to handle the complex, high-value interactions that drive loyalty.

Real estate agencies and property management companies use Eshal to handle the high volume of enquiries around listings, viewings, availability, and tenancy processes. Eshal qualifies leads by gathering requirements upfront, answers property-specific questions instantly, books viewing appointments, and routes serious buyers or tenants to the right agent - so your team spends time on the conversations that convert, not on repetitive information requests.

Insurance organizations use Eshal to handle the high volume of policy, claims, and coverage queries that flood contact centres - instantly answering questions about policy details, claim status, documentation requirements, and renewal processes. Eshal reduces handling time, improves first-contact resolution, and ensures every response is drawn from approved content - reducing compliance risk while improving the customer experience.

Manufacturers use Eshal to support both external customers and internal teams. On the customer side, Eshal handles product specification queries, order tracking, warranty questions, and distributor support. Internally, Eshal gives frontline and field teams instant access to technical documentation, operational procedures, and compliance information - reducing downtime and improving decision-making on the floor.

HR teams use Eshal to give employees instant, self-service access to the information they need - leave policies, benefits, onboarding processes, payroll queries, compliance requirements, and internal procedures. Eshal reduces the volume of repetitive queries reaching HR teams, accelerates onboarding for new employees, and ensures every team member gets consistent, accurate answers regardless of location, shift, or language.

Eshal improves customer service by resolving the majority of routine queries instantly, in the customer's preferred language, at any hour - without making customers wait in a queue or repeat themselves across channels. For the queries that do need a human, full conversation context is handed off automatically, so your team picks up exactly where the AI Concierge left off. The result is faster resolution, more consistent answers, and a support team that can focus on the interactions that genuinely need their judgment.

Yes. Beyond customer-facing use, Eshal is widely used as an internal tool - giving employees instant, self-service answers to IT, HR, and operational questions without needing to file a ticket or wait for a reply. Internal AI Concierges can be deployed for specific teams or departments, each with its own knowledge base and permissions.

IT teams use Eshal to resolve common, repetitive requests instantly - password resets, access requests, software troubleshooting, and policy questions - without a ticket ever reaching a human technician. More complex issues are escalated automatically with full context, so IT staff spend their time on the problems that actually need their expertise rather than the high-volume routine ones.

Sales teams use Eshal to qualify inbound leads instantly, answer common product and pricing questions around the clock, and route qualified prospects directly into the CRM - so reps spend their time on conversations that are ready to close rather than on initial triage.

The best starting point is a product demonstration. Our team will walk you through Eshal's capabilities in the context of your industry and use case, and help you identify where an AI Concierge would deliver the most immediate value. From there, we can scope a proof of concept or move directly to implementation depending on your readiness. Book your demo here.

Implementation involves four key steps: connecting your knowledge sources and enterprise systems, configuring your AI Concierge's persona and conversation flows, deploying across your chosen channels, and testing before go-live. For straightforward deployments, this can be completed in days. For complex enterprise implementations with custom integrations and multiple channels, our team works alongside yours through a structured delivery process.

Eshal supports cloud, on-premise, and hybrid deployment models to accommodate the infrastructure and data residency requirements of different organizations. Cloud deployment is the fastest path to go-live for most businesses. On-premise and hybrid options are available for enterprise and government customers with stricter data governance or sovereignty requirements. Contact our team to discuss the right deployment model for your organization.

Enterprise customers receive a dedicated onboarding experience. This includes a named implementation team, a structured discovery and scoping process, phased rollout planning, knowledge base setup, integration configuration, user training, and a defined go-live process. We work to your timeline and involve the right stakeholders at each stage - IT, operations, customer service, and procurement - to ensure a smooth deployment and rapid time to value.

Our relationship does not end at go-live. Eshal provides ongoing platform support, regular updates, and strategic guidance to help you continuously improve your AI Concierge's performance. Enterprise customers receive dedicated support with defined response times, access to platform expertise, and proactive recommendations based on your usage data. We are invested in your long-term success, not just your initial deployment.

No. Eshal is designed to be configured and managed by non-technical teams - setting up your AI Concierge's knowledge, persona, and conversation flows doesn't require coding or AI expertise. For more advanced enterprise integrations and workflow automation, our implementation team works directly with your IT stakeholders, but day-to-day management of the platform does not require a developer.

Book a demo on our website and we'll set up a walkthrough tailored to your industry and use case. We'll show you how Eshal would work for your specific situation and help you decide whether a proof of concept or a full implementation is the right next step.

Yes. For organizations that want to validate Eshal against a real use case before committing to a full rollout, we offer a structured proof of concept. This typically involves a scoped deployment against a defined set of questions or workflows, allowing your team to evaluate accuracy, integration depth, and fit before scaling to production.

Eshal is built to grow with you. You can start with a single AI Concierge on one channel and expand to multiple Concierges across departments, additional channels, and deeper integrations as your needs evolve - without re-platforming. The same governance, security, and knowledge management foundation supports you whether you're running one use case or dozens across the organization.

No. Eshal's no-code configuration means business teams can build, train, and launch an AI Concierge without writing code or involving a development team. Developers become relevant only when you want deep custom integrations or workflow automation beyond what's available out of the box - and even then, our implementation team handles most of that work directly.

Building your own AI solution gives you full control, but it also means building and maintaining the retrieval pipeline, guardrails, integrations, multilingual support, and security architecture yourself - a significant and ongoing engineering investment. Using a general-purpose tool like ChatGPT or Copilot is fast to start but wasn't built for governed, organization-specific knowledge, enterprise integrations, or the auditability regulated industries require. Eshal sits between the two: enterprise-grade governance, security, and integration depth, delivered as a platform you can configure rather than build - so you get the control of a custom build without the engineering overhead.

After your demo, the typical next step is a discovery conversation to understand your specific requirements, followed - if it's the right fit - by a proof of concept against a real use case. From there, we move into implementation: connecting your knowledge sources and systems, configuring your AI Concierge, and testing before go-live. Once live, our team continues working with you on continuous optimisation as usage data reveals where the AI Concierge can improve further.