Gladly vs Unthread (2026): The CTO's Technical Comparison
A CTO-level comparison of Gladly's B2C person-centric timeline vs Unthread's Slack-native B2B ticketing. Architecture, pricing, APIs, and migration paths.
Planning a migration?
Get a free 30-min call with our engineers. We'll review your setup and map out a custom migration plan — no obligation.
Schedule a free call- 1,500+ migrations completed
- Zero downtime guaranteed
- Transparent, fixed pricing
- Project success responsibility
- Post-migration support included
Gladly is a person-centric, ticketless omnichannel platform built for B2C e-commerce brands that handle voice, email, chat, and SMS at scale. Unthread is a Slack-native conversational ticketing platform built for B2B SaaS and internal support teams that live in Slack. Pick the wrong one and you'll either pay $180+/agent/month for voice capabilities your engineering team never uses, or force your retail customers into a Slack workflow they'll never adopt.
This is not a feature-checklist comparison. You are choosing between two fundamentally opposed data models: a lifelong, ticketless B2C customer timeline versus discrete, channel-based B2B conversational tickets. That one design choice changes routing, analytics, exports, and migration cost.
We've migrated support data across both of these platforms and evaluated their architectures, APIs, and data models at the schema level. This guide is the comparison we'd hand to any CTO choosing between them.
Overview: Two Platforms, Two Completely Different Problems
Gladly was founded in 2014 in San Francisco to replace the ticket-based support model with a person-centric one. Gladly is a customer experience platform founded in 2014 and headquartered in San Francisco. Unlike traditional helpdesk software that creates a new ticket for every interaction, Gladly maintains one continuous conversation per customer across all channels: email, chat, SMS, voice, and social. Its ideal customer is a mid-market to enterprise B2C brand — think Nordstrom, Ulta Beauty, Allbirds — where agents need lifetime customer context across phone, email, chat, and social.
Unthread is a Slack-native (and Microsoft Teams-native) helpdesk that turns chat messages into structured, trackable tickets. Unthread is an agentic service management platform that automates ticket workflows across all enterprise systems. Our conversational AI tracks requests in Slack channels, routes them to agents, and uses automation to resolve tickets before they reach a human agent. We partner with innovative companies like Intuit, Lemonade, and HubSpot to reduce time spent on ticketing by up to 80% — all within Slack. Its ideal customer is a B2B SaaS company using Slack Connect for customer support, or an internal IT/HR/Finance team fielding requests in shared Slack channels.
One-line verdict: Choose Gladly if your support motion is B2C omnichannel with voice. Choose Unthread if your support motion is B2B conversational or internal and your team already lives in Slack.
Core Philosophy & Architecture
Gladly: The Ticketless Person-Centric Stream
Gladly's data model is fundamentally different from traditional helpdesks. There are no ticket numbers. Every customer has a single, lifelong conversation timeline that spans every channel — voice, email, chat, SMS, WhatsApp, Facebook Messenger, and Instagram. Unified Customer View: Gladly provides a single, continuous conversation thread for each customer, allowing agents to see the entire history of interactions across all channels.
This architecture is opinionated: you can't separate "tickets" from the person. An agent sees the entire customer history — including orders pulled from Shopify — the moment a conversation starts. Gladly's routing model assumes one open Conversation at a time and only one assigned agent at a time. That preserves context but creates friction when multiple specialists need to work parallel issues for the same account.
For B2C brands with repeat customers, this is exactly right. If a customer emails about a return on Tuesday and calls about it on Thursday, the agent sees one continuous thread.
For B2B support, this model breaks down. If a single point of contact at an enterprise client emails you about three separate technical issues over a month, Gladly lumps them into one timeline. Separating those issues for SLA tracking or engineering escalation means fighting the platform's native architecture.
For a deeper look at Gladly's setup and configuration, see our Ultimate Guide to Gladly.
Unthread: The Slack-Native Channel-Based Thread
Unthread's architecture starts from the Slack message. Unthread starts from the employee conversation. A request can begin in Slack, Microsoft Teams, email, or a portal, then become a structured ticket with ownership, SLAs, routing, automation, internal notes, and reporting. This is especially useful when employees already ask for help in chat and support teams need those requests to become trackable without adding friction.
Each Slack message (or thread) becomes a ticket in the background. Agents work inside Slack — they don't switch to a separate portal. Unthread adds SLA tracking, assignments, routing, analytics, and AI automation on top of native Slack functionality. Its API centers on conversations with channel IDs, customer IDs, statuses, ticket types, metadata, and messages — a data model built for cross-functional operational workflows, not just branded customer care.
The BYO-LLM Advantage: Unthread supports Bring-Your-Own-LLM architecture via MCP integration. Unthread offers Bring-your-own-LLM capability via MCP integration, avoiding vendor lock-in to specific AI providers. Unthread doesn't train its own models on customer data, and organizations can bring their own LLM via MCP integration to maintain control over AI processing. Enterprises can plug in their own OpenAI or Anthropic API keys, so AI traffic stays within their existing legal agreements rather than flowing through a vendor's shared infrastructure.
Where Each Platform Is Opinionated
| Design decision | Gladly | Unthread |
|---|---|---|
| Primary interface | Custom web app | Slack / Teams |
| Data model | Person-centric timeline (no tickets) | Channel-based threaded tickets |
| Channel strength | Voice, email, chat, SMS, social | Slack, Teams, email, portal |
| AI model | Vendor-managed (Azure OpenAI, Anthropic) | BYO-LLM via MCP |
| Extension model | REST APIs, webhooks, lookup adaptors, commerce integrations | Service-account API keys, webhooks, custom-code automations, MCP |
| Target buyer | VP of CX / Head of Support at B2C brand | Head of IT / Engineering Lead at B2B SaaS |
Feature-by-Feature Comparison
Omnichannel Routing & AI Triage
Gladly routes conversations across voice, chat, email, SMS, and social with native IVR and workforce management. Its AI (Gladly Sidekick) can autonomously resolve common e-commerce requests — order status, refunds, cancellations — by taking direct action on Shopify. Gladly's AI-powered Customer Service Platform fully automates and resolves common customer requests via actions like process return, issue refund, cancel order, and more.
Unthread routes Slack messages to agents based on channel, topic, or AI classification. Danny Fang, Head of IT at Lemonade, describes the impact: "Unthread automatically resolves about 40% of all tickets that come in across different teams. This means countless hours saved for employees across the organization." Its AI is built for technical triage — auto-categorizing Slack messages, drafting responses from engineering documentation, and summarizing long multi-day Slack threads for Jira escalation.
Winner: Gladly for B2C omnichannel. Unthread for Slack-based B2B and internal support. No crossover.
Ease of Use & Learning Curve
Gladly requires onboarding and training — you're adopting an entirely new platform and a new mental model. Agents stop thinking in ticket numbers and start thinking in customer timelines, inboxes, and contacts inside one Conversation. That mental model change is real. The only reason for 4 rather than 5 stars is that we are a small team and Gladly is primarily designed for large / enterprise companies. We've made it work and are happy, but if you're a small team just keep that in mind.
Unthread installs directly into Slack. Initial installation was a breeze and it has a lot of configuration options which makes it easy to customize to our specific organization. Your team doesn't learn a new tool — they keep using Slack. Time-to-value is measured in hours, not weeks.
Winner: Unthread, decisively. Slack-native means near-zero adoption friction.
Customization & Extensibility
Gladly gives you app integrations, lookup adaptors, webhooks, and a developer platform built to surface external customer data and actions inside the agent's profile view. The core workflow is deliberately rigid to preserve the timeline experience. You can inject custom cards via the API, but you're not rebuilding the UI.
Unthread operates within Slack, meaning its UI is Slack's UI. It offers a modern REST API, service accounts, webhooks, custom-code automations, and BYO AI options on higher plans. You can trigger complex webhooks, sync with engineering trackers, and build custom Slack workflows that interact directly with Unthread's ticketing layer.
Winner: For commerce-heavy data enrichment, Gladly has the cleaner story. For workflow scripting and developer-led automation, Unthread is more flexible.
Collaboration
Unthread supports triage threads, followers, collaborators, user-group assignment, rotations, approval requests, project-level visibility, and an inline audit trail. This makes it strong when multiple specialists need to work parallel sub-issues — the exact pattern in B2B technical support.
Gladly is optimized for single-owner customer care. One open Conversation, one assigned agent at a time. That works well for B2C interactions where you want clear ownership, but limits workflows where engineering, product, and support all need hands on a complex account issue.
Winner: Unthread for multi-team collaboration. Gladly for clear single-owner customer relationships.
Detailed Comparison Table
| Capability | Gladly | Unthread | Winner |
|---|---|---|---|
| Omnichannel (voice, email, chat, SMS, social) | Native across all channels | Slack, Teams, email, portal | Gladly |
| Voice / IVR support | Built-in with call recording, AI voice | Not applicable | Gladly |
| Slack-native ticketing | Not supported | Core architecture | Unthread |
| AI auto-resolution | E-commerce actions (refund, cancel, order status) | Internal IT/HR workflows, ~40% auto-resolve | Depends |
| BYO-LLM / data privacy | Vendor-managed LLMs | BYO-LLM via MCP integration | Unthread |
| E-commerce integrations | Shopify, BigCommerce native | None | Gladly |
| Engineering tool integrations | Limited | Jira, Linear, PagerDuty native | Unthread |
| Knowledge base | Gladly Answers (public + internal) | Self-learning KB from resolved tickets | Tie |
| SLA enforcement | Built-in | Built-in with Slack-native alerts | Tie |
| Reporting & analytics | Extensive reporting suite with exports | Custom dashboards, CSAT/NPS surveys | Gladly |
| Collaboration model | Single-owner per conversation | Multi-team triage, rotations, approvals | Unthread |
| Learning curve | Weeks (new platform + mental model) | Hours (Slack-native) | Unthread |
| Minimum commitment | 10 seats, annual contract | 5 seats | Unthread |
API Limits, Performance & Scalability
When evaluating helpdesks at enterprise scale, API rate limits are the silent killers of operational efficiency — especially during migrations and custom integrations.
Gladly's API Constraints
Gladly's REST API uses Basic Auth and enforces strict rate limits:
| API Type | Request Limit | Concurrency Limit |
|---|---|---|
| Standard (GET/POST/PUT/PATCH/DELETE) | 10 requests per second | N/A |
| Reporting APIs | 10 requests per minute | 2 concurrent requests per org |
The conversations endpoint returns at most 100 conversations per customer (unpaginated), and the timeline endpoint returns at most 1,000 items per conversation. If a customer or conversation exceeds these limits, the Gladly-Limited-Data response header flags the truncation — but you can't paginate past it through the standard API.
For data extraction, Gladly offers a file-based Export API that generates JSONL files (agents, customers, conversation_items, topics). Export jobs older than 14 days are automatically removed, and exports spanning more than six months may incur a service fee. Production reporting data runs about one hour behind live.
Gladly's Reporting API limit of 10 requests per minute with only 2 concurrent requests is the primary bottleneck during migrations. Any automated extraction must implement exponential backoff with jitter to avoid 429 errors. Hitting the Reporting API with a standard polling script will immediately break your analytics pipelines.
Unthread's API & Scale Characteristics
Unthread's architecture is inherently tied to Slack's API infrastructure. Your scale ceiling is partly governed by Slack's own rate limits, which vary by endpoint and tier. The platform supports SSO integration, offers isolated hosted environments for enterprise customers, and works with Slack Enterprise Grid.
Unthread's own API is lightweight — primarily used for syncing metadata, managing conversations, and triggering external workflows. The practical operational caveat: automatic ticket tracking only works when the Unthread app is installed in the channel, the right channel is connected, and an external user posts a message. Some Slack Connect channels created by another workspace may not be connectable without the creator changing permissions.
Teams support is still in beta and does not support DMs or autoresponses. For large organizations, Unthread supports Slack Enterprise Grid — the multi-workspace architecture that large enterprises use.
Pricing & Total Cost of Ownership
Gladly Pricing
Gladly does not publish pricing on its website. Based on verified buyer data:
Based on verified reports from customers and pricing intelligence platforms, you're looking at roughly $180 to $210 per user per month with annual contracts and minimum seat requirements. For a 10-person team, that's $21,600 per year minimum.
Starting price: ~$180–$210 per agent per month (annual contracts). Pricing structure: Custom pricing with minimum seat commitments. Extra costs: Voice minutes, SMS usage, add-ons, and professional services. Cost predictability: Low. Spend increases as usage and headcount grow. Best fit: Large B2C teams where voice is the primary support channel.
The Superhero tier (for large enterprises) runs approximately $210 per user per month and requires a minimum of 45 users. That's over $113,000/year before usage-based charges for voice and SMS.
Gladly typically requires a 12-month commitment, with billing handled upfront or yearly rather than monthly. Many contracts include a minimum number of agent seats, so teams pay for a fixed count even if headcount drops.
Unthread Pricing
Unthread publishes pricing transparently:
The 'Basic' plan at $50 per agent/month is geared toward small teams needing to move beyond disorganized Slack threads into structured ticketing with SLAs. The 'Pro' plan at $75 per agent/month adds AI automation builder, self-learning documentation, CRM integrations, and custom analytics dashboards. Enterprise pricing is custom and includes HIPAA BAAs, dedicated hosting, SSO, and Slack Enterprise Grid support.
All plans require a minimum of 5 seats and include a 14-day free trial. Contract lengths can be monthly, annual, or multi-year. Critically, Unthread's model doesn't punish cross-departmental collaboration — engineers and product managers can participate on tickets without each requiring a dedicated seat at enterprise-tier pricing.
TCO Comparison
| Scenario | Gladly (annual) | Unthread (annual) |
|---|---|---|
| 5 agents | Not available (below 10-seat minimum) | $3,000 (Basic) / $4,500 (Pro) |
| 10 agents | ~$21,600+ | $6,000 (Basic) / $9,000 (Pro) |
| 25 agents | ~$54,000+ | $15,000 (Basic) / $22,500 (Pro) |
| 50 agents (enterprise) | ~$108,000–$126,000+ | Custom (Enterprise tier) |
Gladly figures exclude voice minutes, SMS usage, and professional services. Unthread figures exclude Slack subscription costs.
Don't forget Slack's own per-user cost when budgeting Unthread. On Slack Pro, that's roughly $8.75/user/month — which adds up for large organizations. On Slack Enterprise Grid, pricing is custom.
Security, Compliance & Data Sovereignty
| Security area | Gladly | Unthread |
|---|---|---|
| SOC 2 Type II | Yes | Yes, with regular independent audits and penetration testing |
| PCI-DSS | Yes, PCI-DSS compliant for payment data processing | Not applicable |
| HIPAA / BAA | Not publicly advertised | Available on Enterprise plans |
| GDPR | DPAs signed with all third-party vendors | Supported |
| BYO-LLM | No — uses vendor-managed LLMs | Yes, via MCP integration |
| Encryption | AES-256 at rest, TLS 1.2 in transit | AES-256 at rest, TLS 1.3 in transit |
| Data residency | Cloud-hosted (US) | Isolated hosted environments on Enterprise (US, EU, UK, Australia, Canada) |
| SSO | Supported | Supported |
| Audit logs | Available, with compliance event log for deletions and redactions | Available, with IP and geo metadata |
For regulated industries: Gladly's PCI-DSS compliance makes it the stronger choice for B2C brands processing payments inside support conversations. Unthread's BYO-LLM and HIPAA BAA make it stronger for enterprises with strict data processing requirements — your AI traffic never touches a third-party vendor's infrastructure outside your own agreements. For data sovereignty and internal regulated workflows, Unthread is easier to underwrite from public documentation.
Integrations & Ecosystem
Your helpdesk must map to your system of record.
Gladly integrates deeply with e-commerce stacks. Gladly focuses heavily on integrations with major e-commerce platforms like Shopify and BigCommerce. The Shopify integration is built and maintained by Gladly, enabling agents to view orders, process refunds, and issue discount codes without leaving the platform. It also connects to loyalty platforms (Yotpo, LoyaltyLion), CRM systems, and workforce management tools. If you need agent-facing order actions and customer enrichment, Gladly has the cleaner story.
Unthread integrates with engineering and ops tools — Jira, Linear, GitHub, PagerDuty, Salesforce, HubSpot CRM — plus bi-directional ticket sync with external helpdesks. Pro plan includes bi-directional 3rd party ticket sync, CRM integrations, and advanced permissioning. Its task-manager sync between Jira issues and Slack triage threads is exactly the bridge B2B support teams need: a bug reported in a Slack channel gets linked to a Linear issue, and the Slack thread automatically updates when the engineering team merges the fix.
Winner: Gladly for e-commerce context. Unthread for engineering-centric B2B workflows.
Migration & Lock-in: How to Transition Safely
Switching between Gladly and Unthread isn't just a data migration — it's a data model translation. You're mapping a person-centric timeline (no ticket IDs, conversations grouped by customer identity across channels) to channel-based Slack threads (or vice versa). This is a fundamentally different schema, not a field-mapping exercise.
Gladly customer profile + lifelong conversation
→ one Unthread customer
→ one or more Unthread conversations by channel or thread
→ internal notes mapped to private notes or triage messages
→ SLA and assignee history preserved as metadata or custom fieldsGetting Data Out of Gladly
Gladly offers an Export API that produces JSONL files for agents, customers, conversation_items, and topics. The export covers all delivered communications within a date range. Files older than 14 days are purged, and exports spanning more than six months may incur a service fee. All-time or multi-day historical raw transcript exports may require Gladly Support involvement.
The standard REST API caps at 100 conversations per customer and 1,000 items per conversation — unpaginated. For high-volume customers, this means data truncation unless you use the Export API.
Getting Data Out of Unthread
Unthread's data lives partly in Slack (message content, threads, reactions) and partly in Unthread's own system (tickets, SLAs, analytics, internal notes). Conversations, messages, customers, files, reporting, and webhooks are all exposed via the API, making it more straightforward to script. The lock-in risk is less about export format and more about operating model: Slack thread URLs, channel IDs, project visibility, and Slack Connect topology don't automatically translate to another helpdesk.
Why DIY Migration Fails Here
The architectural mismatch makes CSV-based migration impractical. Gladly's timeline model has no concept of discrete tickets to export row-by-row. Unthread's threaded conversations don't map cleanly to Gladly's person-centric stream. Internal notes, SLA metadata, and custom attributes have no 1:1 equivalents.
If you are migrating out of Gladly, you face a severe data engineering challenge. You must export massive continuous JSON timelines for every customer and programmatically chunk them into logical, discrete tickets based on time decay or topic shifts — before importing into any standard helpdesk. Attempting this with CSV exports typically results in lost context, broken threading, or orphaned records.
This is where ClonePartner fits. We automate the complex schema translation between these fundamentally different data models — mapping Gladly's lifelong person-centric timelines into Unthread's threaded tickets (or the reverse) without losing conversation history, internal notes, or metadata. Our scripts handle Gladly's 10 req/sec API limits with built-in backoff, and we run zero-downtime migrations so your team never stops serving customers during the transition.
For guidance on what data to migrate versus what to archive, see our Helpdesk Migration Playbook.
Use-Case Recommendations
Choose Gladly If:
- You're a B2C e-commerce brand with Shopify or BigCommerce and agents who handle voice calls, emails, and chat simultaneously
- You need native voice/IVR — Gladly's built-in telephony with AI voice is hard to match
- Repeat customer relationships matter — the person-centric timeline gives agents instant context across months or years of interactions
- Your budget supports $180+/agent/month and your team has 10+ agents
- PCI-DSS compliance is required for in-conversation payment processing
Choose Unthread If:
- Your team lives in Slack and your support motion is B2B shared channels or internal IT/HR/Finance requests
- You're a startup or mid-market team with 5–25 agents who need structured ticketing without a heavy platform
- Data privacy is non-negotiable — BYO-LLM means your AI data never leaves your own agreements
- You need fast time-to-value — Unthread installs in hours, not weeks
- Budget is a constraint — at $50–75/agent/month, Unthread is 60–70% cheaper than Gladly
- HIPAA compliance is needed for internal support workflows
Industry-Specific Guidance
| Scenario | Recommendation |
|---|---|
| DTC e-commerce, 15+ agents, voice-heavy | Gladly |
| B2B SaaS, shared Slack channels with customers | Unthread |
| Internal IT helpdesk (any size) | Unthread |
| Hospitality/travel, multi-channel | Gladly |
| Startup, <10 agents, budget-conscious | Unthread |
| Enterprise retail, 50+ agents | Gladly |
| Regulated industry, HIPAA for internal workflows | Unthread |
| Regulated industry, PCI-DSS for payments | Gladly |
Strengths & Weaknesses
Gladly
Strengths:
- The person-centric timeline is genuinely superior for B2C brands with repeat customers
- Native voice/IVR with AI is best-in-class for support teams that handle calls
- Deep Shopify integration lets agents and AI take direct action on orders
- PCI-DSS compliance for in-conversation payments
- Proven at enterprise scale — Nordstrom, Ulta Beauty, Crate & Barrel
Weaknesses:
- For small teams or startups, Gladly is likely overkill. The $21,600 minimum annual commitment puts it out of reach for many growing companies.
- Adding or removing agents mid-contract often requires renegotiation, which further reduces cost predictability as team size changes.
- You have to go all-in on the Gladly ecosystem. If your team is happy with its current help desk or uses specific tools for different channels, moving everything over to Gladly is a huge project. This all-or-nothing approach can feel pretty rigid.
- One-Conversation, one-owner routing doesn't fit B2B multi-team workflows
- API pagination gaps (100 conversations max, 1,000 items max) create extraction challenges for migrations and BI
Unthread
Strengths:
- Slack-native architecture means near-zero adoption friction and fast deployment
- BYO-LLM keeps AI data within your own legal agreements
- Transparent, predictable pricing at 60–70% lower cost than Gladly
- Self-learning knowledge base automatically generates help articles from resolved tickets
- Unthread serves as a source-of-truth ticketing system where automation ensures every customer interaction automatically creates a ticket so nothing slips through.
Weaknesses:
- No voice/IVR support — if phone is a primary channel, Unthread is not an option
- No native e-commerce integrations (Shopify, BigCommerce)
- Fully dependent on Slack's infrastructure — if Slack goes down, Unthread goes down
- Teams support is still in beta and limited (no DM support, no autoresponses)
- Ticket tracking can fail if Slack channel and app permissions are misconfigured
The Verdict
The decision between Gladly and Unthread is binary, based on your support motion.
Choose Gladly if you're a B2C brand with 10+ agents, a voice-heavy support motion, and the budget for a $180+/seat enterprise platform. You'll get a person-centric timeline, deep Shopify integration, and native telephony that no Slack-native tool can replicate.
Choose Unthread if your support is B2B conversational (Slack Connect shared channels) or internal (IT, HR, Finance requests), you need transparent pricing under $75/agent/month, and you want your AI data to stay within your own infrastructure via BYO-LLM.
These two platforms solve fundamentally different problems. The right choice depends entirely on your support motion, not a feature checklist. Gladly is overhyped when sold as a drop-in replacement for standard ticket queues — the data model change is bigger than vendors admit. Unthread is underrated when the real job is Slack-based support, internal service delivery, or B2B shared channels.
The bigger risk isn't choosing the wrong feature set — it's underestimating the switching cost between two very different data models. Migrating between a ticketless person-centric model and a channel-based threading model requires schema translation, not data copying. ClonePartner handles exactly this: automated mapping between Gladly and Unthread data models, built-in rate-limit handling for Gladly's API constraints, preservation of conversation history and internal notes, and zero downtime throughout the transition.
Frequently Asked Questions
- Can Gladly and Unthread be used together?
- Technically yes — some organizations run Gladly for external B2C support and Unthread for internal IT requests — but there's no native integration between them. You'd need custom webhooks or a migration service like ClonePartner to sync data between the two platforms.
- What's the biggest migration risk between Gladly and Unthread?
- The data model mismatch. Gladly stores conversations as person-centric timelines without ticket numbers. Unthread stores conversations as channel-threaded tickets. Mapping one to the other requires schema translation — not just field mapping — and mishandling it results in lost conversation context, broken threading, or orphaned records.
- How do Gladly's API rate limits affect integrations?
- Gladly enforces 10 requests per second for standard API operations and 10 requests per minute (with 2 concurrent requests max) for Reporting APIs. The conversations endpoint also caps at 100 results per customer, unpaginated. Any integration or migration script that doesn't implement exponential backoff with jitter will hit 429 errors quickly.
- What does BYO-LLM mean in Unthread?
- Bring-Your-Own-LLM allows enterprises to connect their own OpenAI or Anthropic API keys to Unthread via MCP integration. This ensures that sensitive data processed by AI stays governed by your existing enterprise agreements rather than flowing through a third-party vendor's shared infrastructure.
- Does Unthread support HIPAA compliance?
- Yes. Unthread offers HIPAA compliance with Business Associate Agreements (BAAs) on its Enterprise plan, along with isolated hosted environments and BYO-LLM to keep data within your own infrastructure.
