Embedded analytics cost: what you pay for when the software is free
Apache Superset has no paid edition, so its license cost is zero, but the bill still arrives through other lines. A product or portal team comparing Superset, Metabase, a partner build and commercial BI embedding should expect to pay for four things: the infrastructure that runs Superset, the analytics database that answers its queries, the embedding and row-level security work that makes dashboards safe for each customer, and the upkeep that keeps the stack patched. Superset's architecture documentation describes a metadata database, an optional caching layer and optional Celery workers, and alerts, async queries and thumbnails depend on the caching and worker components. Each of those is a server someone must size, back up and operate. Superset is licensed under Apache-2.0 (see the Apache Superset repository), so the recurring cost is people and hosting, not seats.
Metabase and commercial BI shift the cost toward licenses. Metabase's open source edition is free, but the pricing page places SSO, row and column level permissions and multi-tenant data segregation in its Pro and Enterprise plans, and those plans charge per user beyond the included seats. Commercial BI embedding is usually quoted by sales or by capacity, so the reader should ask for the embedded tier, the user or session model and the renewal terms in writing. The honest estimate depends on four inputs: the number of external users and internal staff who will log in, the tenant isolation model (shared database with row-level filters, or one database per customer), whether SSO or guest-token embedding is required, and data volume and query frequency, which determine how large the analytics database must be.
Before comparing totals, have each option quoted against the same four inputs and ask the same questions: which features require a paid tier, what the one-time setup covers, what recurs each year, and who carries upgrades and restore testing. The reader should also confirm that row-level security is enforced by the platform or in the query layer, because that work is the largest hidden cost in any build. DeployOpen does not publish prices, but a fixed-scope written proposal can show the infrastructure, analytics database, embedding and security work, and upkeep as separate lines so that open source and licensed options can be compared on equal terms.
Cost components by build, partner and SaaS option
This table compares the cost components of building in-house on Apache Superset or Metabase open source, engaging an implementation partner, and buying commercial SaaS or embedding. Read each row for who is paid, whether the cost is one-time or recurring, and what to ask for in a written quote. The main takeaway is that open-source licenses remove the license line but not the engineering, security and upgrade work.
| Cost component | Build in-house (open source) | Implementation partner | Commercial SaaS or embedding |
|---|---|---|---|
| Software license | Apache Superset carries no paid edition under Apache-2.0. Metabase Open Source is free, but SSO and row-level permissions need Pro or Enterprise per the Metabase pricing page. | Partner fees cover the work, not the license. Confirm which Metabase tier the scope assumes and who holds the subscription. | Metabase Starter is $100 per month or $90 billed yearly with 5 users, and Pro is $575 per month or $517.50 billed yearly with 10 users (Metabase pricing page). Power BI Premium Per User and Looker Standard, Enterprise and Embed are priced per the Microsoft and Google Cloud pages. |
| Infrastructure and analytics database | Your team pays cloud or data center costs for app servers, a metadata database, caching and workers, plus the analytics warehouse. Recurring. | Often billed as a separate infrastructure line on the client's account. Confirm who owns the cloud account and who pays the bill. | Usually included in the subscription for the BI layer, but the analytics database may still be your cost. Confirm whether query compute is metered. |
| Embedding and row-level security build | One-time engineering to map tenants to roles, guest tokens or signed embedding, and test that no customer sees another's rows. Your staff are paid. | Fixed-scope work, one-time, covering embedding, row-level filters and tests. Ask for test evidence in the quote. | Vendor provides embedding SDKs or signed embedding (for example, Looker Embed). Your integration work remains, and per-viewer or capacity fees may apply. Ask how each is metered. |
| SSO and identity integration | One-time setup of OAuth2, OIDC, SAML or LDAP group mapping. Metabase SAML, JWT and SCIM need Pro or Enterprise per the Metabase people and groups docs. | One-time configuration, often included in the fixed scope. Confirm which identity provider and plan tier are assumed. | Typically included in higher tiers only. Confirm which SSO protocols and group sync features sit in the quoted edition. |
| Upgrades, backups and support | Recurring staff time for upgrades, metadata database backups and staging tests. Superset requires running superset db upgrade and superset init after backing up. | Available as a recurring retainer after go-live. Ask for upgrade cadence, staging tests and restore tests in writing. | Vendor handles platform upgrades, but your integration code and embedding changes still need regression testing. Confirm the support tier. |
| Enterprise or high-volume scale | Costs rise with query load, which needs a separate replica or warehouse to keep production databases safe. Recurring. | Scope a sizing and load test as a one-time item. Ask what changes at higher user counts. | Enterprise pricing is by quote, with Metabase Enterprise starting at $20,000 per year (Metabase pricing page). Power BI Embedded is quoted by sales. |
| Who gets paid | Your internal team and cloud providers. License fees are zero for Superset and Metabase Open Source. | Your partner for fixed-scope work, plus cloud providers and any licensed BI vendor you choose. | The SaaS vendor for subscriptions and usage, plus your team for integration and identity work. |
Where the money goes in a self-hosted Superset or Metabase deployment
The recurring cost of a self-hosted Superset deployment sits in its components, not in the license, because Apache Superset is Apache-2.0 licensed and has no paid edition. A production stack has a Flask and React application, a metadata database that must be PostgreSQL or MySQL (the SQLite default of Quickstart and PyPI installs is not recommended for production), an optional caching layer, and optional Celery worker and beat processes. Alerts and reports, async queries and dashboard thumbnails depend on the caching layer and workers, and most installations use Redis and Celery for them. Each of these is infrastructure you or your partner must size, patch, monitor and pay for. Confirm the supported metadata database versions (PostgreSQL 10 to 17, MySQL 5.7 and 8.x) against your chosen release before you price the database tier.
The analytics database is usually the largest component to plan. ClickHouse is open source under Apache-2.0, but ClickHouse Inc. sells ClickHouse Cloud separately, so the self-hosted option moves the operating work to your team. Its replication guide recommends at least three ClickHouse Keeper nodes on dedicated hosts in production, with about 4 GB of RAM each, and replication alone does not protect against human error or corruption. Backups through BACKUP and RESTORE need storage you pay for, and the backup documentation warns that a backup never test-restored will probably not restore when needed, so restore tests on a spare cluster are a recurring task. Ask your partner to price these as separate line items: Keeper hosts, backup storage and restore drills.
Security and upkeep also recur. Superset requires a user-specified SECRET_KEY to start, and it supports rotation through PREVIOUS_SECRET_KEY, which someone must schedule and document. Its upgrade guidance says to back up the metadata database, run superset db upgrade and superset init, test in staging first and check UPDATING.md for breaking changes. Those steps are labor every release cycle, whether an internal platform team or an implementation partner performs them under a retainer. When you compare quotes, ask who owns the staging environment, who reviews breaking changes, and how long a rollback is expected to take.
What moves the embedded analytics cost up or down
Embedded analytics cost is set less by the software license than by who sees the dashboards, how tenants are separated, and how much operational work the platform creates. The cards below cover the decisions that change the total, and each one is a question to settle in the written scope before a quote is compared.
Named users and capacity pricing
Commercial BI tools price by seat, by capacity or by query allowance, so the number of end users and the model you choose can change the annual total more than any single feature. Metabase Starter includes 5 users and Pro includes 10 users, with additional users billed per month, while Power BI Premium Per User and Fabric capacity are priced on different axes. Ask the vendor how external customers are counted, whether guest or anonymous viewers are billable, and how the price changes when the audience doubles. Model the audience before comparing vendors.
Tenant isolation and row-level security
Each customer or portal user must see only their own rows, and that requirement often decides which plan or build path is viable. In Metabase, row and column level permissions and multi-tenant data segregation sit in the Pro and Enterprise plans, so an open source deployment must either enforce isolation at the database or accept the paid tier. Superset supports row-level security without a paid edition, but the design, testing and ongoing review of those rules is labor that a partner will price. Confirm which layer enforces isolation and who tests it after each change.
SSO and group mapping requirements
Single sign-on and group-to-role mapping are the most common reason a Metabase deployment moves from the open source edition to a paid plan. JWT, OIDC, SAML and SCIM provisioning require Pro or Enterprise, while Google sign-in and basic LDAP are available on Starter. Superset maps OAuth or LDAP groups to roles with AUTH_ROLES_MAPPING and supports Google and Azure, so its identity cost shifts to configuration and testing effort. List your identity provider, the group model and the provisioning expectations before deciding which product fits.
Data volume and query load on production
Dashboards that run against transactional databases compete with the application for capacity, and that competition is paid for in database sizing, replicas or a separate analytics store. ClickHouse is open source under Apache-2.0 and suits high-volume analytical queries, but it adds its own cluster, Keeper nodes and backup storage to the bill. Estimate concurrent viewers, refresh frequency and the largest tables before choosing infrastructure. Ask any partner how query load is isolated from production systems and what sizing assumptions the quote uses.
High availability and backup targets
Replication does not protect against human error or corruption, so a credible availability design includes tested backups as well as replicas. ClickHouse recommends at least three Keeper nodes on dedicated hosts in production, and its backup documentation warns that a backup never test-restored will probably not restore when needed. Superset requires a metadata database, and production installs typically add a caching layer and workers for alerts, reports and thumbnails. Each added node, restore test and staging environment is recurring infrastructure and operations cost, so set the recovery targets in writing before comparing quotes.
Upgrade cadence and version support
Upgrades are a recurring cost that scales with how many components you run and how often they change. Metabase LTS lines are supported for 14 months, non-LTS releases usually reach end of life after 2 months, and Metabase upgrades run application database migrations that require a backup and a single node for major versions. Superset upgrades need a metadata database backup, superset db upgrade and superset init, plus a check of UPDATING.md. ClickHouse stable releases ship roughly monthly while LTS releases ship twice a year. Choosing an LTS line for each product and budgeting staging test cycles keeps this cost predictable.
Build, partner or SaaS: when each option carries the lower total cost
An in-house build fits teams that already employ platform engineers and security engineers who can own the analytics stack after launch. With Apache Superset, the software carries no license fee because the project is released under the Apache License 2.0 and has no paid edition, so the cost sits in people, infrastructure and the analytics database. Most of the spend is recurring: staff time for upgrades, security patching, tenant isolation testing and on-call support.
- Cost pattern: Little or no license spend, but a steady share of engineering time for each upgrade cycle and each new embedding or permission change.
- Budget risk: Staff turnover can leave one engineer holding the only working knowledge of row-level security rules and the embedding configuration.
- Hidden work: Superset needs a metadata database, and alerts, async queries and thumbnails need a caching layer and workers, so the team must operate more components than a single application.
- Upgrade discipline: The upgrade documentation calls for a database backup, a staging test and a review of UPDATING.md before each version change.
- Confirm first: Ask whether your engineers have capacity for this work each quarter, and confirm the supported metadata database versions against the configuration documentation before you commit.
Build is rarely cheapest when the team is small or when the first customer-facing release is due on a fixed date. Estimate the recurring engineering load before comparing it with a quote.
An implementation partner delivers a fixed-scope project in which scope, deliverables and timelines are agreed in writing before work starts. The client keeps the infrastructure, data and open-source licenses, so the organization pays for the delivery of a defined outcome rather than a seat count. Cost is front-loaded into design, build, security hardening and handover, followed by an optional retainer for upgrades and support.
- Cost pattern: A one-time project fee tied to the written scope, plus any ongoing support or upgrade retainer agreed separately.
- Budget risk: Scope creep, such as extra data sources, additional tenants or new theming requirements, can move the final cost if the statement of work is vague.
- Infrastructure stays with you: Cloud accounts, the analytics database and backups remain in your environment, so those recurring costs are yours to plan.
- Handover: Ask what documentation, runbooks and training the partner delivers so that your team can operate the stack without them.
- Confirm first: Ask for the deliverables list, the acceptance criteria, the support hours included in any retainer and how upgrades are priced. DeployOpen does not publish a price list, so request a quote built from the written scope.
This model suits organizations that want a defined outcome without hiring a platform team for the build, though it still requires internal ownership after handover.
Commercial BI embedding moves hosting, upgrades and most security operations to the vendor, and the cost is paid as a subscription. Pricing typically follows users, capacity or query allowances, and enterprise embedding tiers are often quote-based. Data may be processed in the vendor's environment, so data location and processing terms become part of the budget conversation.
- Cost pattern: Recurring subscription that grows with users or capacity. Metabase lists Starter at $100 per month or $90 per month billed yearly with 5 users included, and Pro at $575 per month or $517.50 per month billed yearly with 10 users included, with extra users at $6 or $12 per month (Metabase pricing page). Looker's Embed edition and Power BI Embedded are quoted by sales.
- Feature gating: On Metabase's pricing page, SSO, row and column level permissions and multi-tenant data segregation appear only in Pro and Enterprise, so the security features you need may set the plan.
- Budget risk: Per-user pricing can rise sharply when a customer-facing audience grows, and capacity or API allowances can be exceeded during reporting peaks.
- Data processing: Confirm where dashboard data is processed and stored, and what sub-processors the vendor uses, before you embed customer data.
- Confirm first: Ask for the total annual cost at your projected viewer count, the embedding license terms, the export and exit provisions and the feature list for your chosen tier.
SaaS usually has the lowest start-up effort, but the recurring cost and data location terms should be modeled over several years.
How to build your own embedded analytics cost estimate
This sequence turns an embedded analytics idea into a cost estimate you can compare across build, partner and SaaS options, and it ends with written quotes that can be compared line by line.
Count internal and external users and tenants
Start by counting who will see dashboards: internal staff, customers, partners and the number of separate tenants or client organizations. Record how many people log in each month, how many are active at peak reporting times, and whether any viewers need access without a login. Pricing on per-user plans, such as Metabase's Starter and Pro tiers, depends on these counts, so the numbers drive the license line. The step is done when you have a written table of user groups, tenant counts and peak concurrency that the product owner and security lead have both signed off.
List required SSO and row-level security features
Write down how each viewer proves who they are (SSO through your identity provider, OAuth, LDAP or guest tokens) and what each viewer may see at the row level, such as a single client, facility or account. Then check which of these features your candidate tier includes. The Metabase pricing page places SSO, row and column level permissions and multi-tenant data segregation in its Pro and Enterprise plans, so a requirement here can move you to a different price line. Confirm each feature with the vendor in writing. The step closes when every requirement is marked as included, excluded or to be built.
Choose the license tier or open source path
Decide whether the open source edition meets your requirements or whether a paid tier or commercial BI embedding is needed. Apache Superset has no paid edition, so its cost sits in infrastructure and implementation. Metabase Open Source is free to use, but the AGPL terms and the missing SSO and row-level features shape the choice. Commercial options such as Looker or Power BI Embedded are quoted or priced by their own schedules, so collect the current pricing pages and any sales quote. Record the license terms that govern distribution to your customers, not only your internal use. The step is complete when one license path is selected with a written reason.
Size the analytics database and cache
Estimate the data volume, query frequency and refresh schedule that the dashboards will generate, and decide where the analytical data lives. ClickHouse is open source under Apache-2.0 and is often used for high-volume analytical queries, but it still needs compute, storage and replicas sized to your workload. Superset's alerts, async queries and thumbnails need a caching layer and Celery workers, and most installations use Redis and Celery for them. Ask your engineers to run representative queries against a copy of the data before sizing. The step is done when you have a sizing note covering compute, storage, cache and replica counts.
Budget infrastructure in your own environment
Price the servers, Kubernetes or virtual machine capacity, storage, load balancers, backups and monitoring in the cloud account or data center where the platform will run. Include the metadata database, which must be PostgreSQL or MySQL for production Superset or Metabase, and the replicas needed for high availability. Split these costs into one-time setup and recurring hosting, and note which items scale with users, queries or data volume. Ask your cloud or infrastructure team to price these against your existing contracts. The step closes with a monthly and annual infrastructure estimate that lists every component.
Add embedding and security work as one-time costs
Estimate the one-time effort for embedding dashboards in your product or portal, configuring SSO and row-level security, testing tenant isolation, setting up audit logging and hardening the platform. Each item is paid to your own engineers or to an implementation partner, so the quote should name who does the work and what is delivered. Ask for a fixed-scope statement that lists the embedding method, the identity integration, the tests and the documentation. The step is complete when each security and embedding item has a scope line and an owner.
Add recurring upgrades, backups and restore tests
Include the ongoing work that keeps the platform secure and recoverable. Superset upgrades require backing up the metadata database, running database upgrades, testing in staging and checking the upgrade notes for breaking changes. Metabase upgrades run application database migrations automatically, and its LTS versions are supported for 14 months. ClickHouse backups need regular restore tests, because a backup never test-restored will probably not restore when needed. Estimate the recurring effort for these tasks in your team or in a retainer. The step is done when a recurring operations line exists for each platform component.
Request written quotes with scope and exclusions
Ask every vendor or implementation partner for a written quote that states scope, deliverables, acceptance criteria, what is excluded and what is recurring versus one-time. Compare the quotes against your sheet line by line, and flag any item that appears in one quote but not another. DeployOpen does not publish a price list and quotes follow from written scope, so ask for the same scope in each response. Confirm license fees, support terms and hosting assumptions in writing. The step closes when the quotes are normalized into one comparison that the budget owner has reviewed.
Why the license line can change the whole budget
The license on the tool you pick sets the floor of the budget before any engineering starts. Apache Superset is licensed under Apache-2.0 and has no paid edition, so the software line is zero. Metabase is different. Its Open Source edition is licensed under the AGPL, and the controls a customer-facing product usually needs sit behind paid plans. The Metabase pricing page places single sign-on, row and column level permissions, multi-tenant data segregation and usage analytics and auditing in the Pro and Enterprise plans only, not in Open Source or Starter. Pro is listed at $575 per month, or $517.50 per month billed yearly, with 10 users included and $12 per month for each additional user. Metabase Enterprise is quoted by sales starting at $20,000 per year.
Those numbers only mean something once you map your requirements to features. If your product needs per-customer row-level security and SSO from your identity provider, the Open Source edition does not cover them, so the choice is between pricing a paid tier at your expected user count or building those controls yourself on Superset or Metabase Open Source. The second option moves the cost into engineering time, testing and ongoing maintenance, which recur with every upgrade. Ask each candidate partner to list, feature by feature, what the quoted scope includes and which items depend on a paid plan or custom code. Also check how the per-user pricing scales if your external user count grows from hundreds to thousands.
Distribution terms are the other half of the license question. The AGPL obligations that apply when you modify Metabase and offer it to users over a network are covered in depth on the SaaS embedding topic, but the cost consequences belong here. A change in how you distribute or modify the tool can change whether you owe source disclosure or need a commercial license, and that can change the budget more than any plan fee. Confirm the obligations for your specific architecture with counsel before you fix the budget, and record their answer in the scope document so the quote and the legal position match.
Questions buyers ask about embedded analytics cost
How much does it cost to run Apache Superset?
Superset has no paid edition, so the software license cost is zero, and the recurring cost is the infrastructure and people who run it. The Superset architecture documentation lists four components: the Flask and React application, a metadata database, an optional caching layer and optional Celery worker and beat processes. Alerts, reports, async queries and dashboard thumbnails need the caching and worker components, and most installations use Redis and Celery for them. Production runs on PostgreSQL or MySQL, not the SQLite default. Ask a partner to quote the hosting, the analytics database that serves your queries, backups, upgrades and on-call support as separate line items.
Is Metabase free enough for embedding in a product?
The Open Source plan is free, but it does not include the controls most embedded products need. According to the Metabase pricing page, SSO, row and column level permissions, multi-tenant data segregation and usage analytics and auditing are included only in the Pro and Enterprise plans. Starter is $100 per month or $90 per month billed yearly with the first 5 users included, and Pro is $575 per month or $517.50 per month billed yearly with the first 10 users included, with additional users at $6 (Starter) or $12 (Pro) per month. Confirm the current terms on the pricing page and check the AGPL obligations with counsel before you distribute a modified build.
What does a partner implementation include and how is it quoted?
A partner implementation covers the work your team would otherwise absorb: planning the architecture, deploying the platform in your cloud account or data center, configuring authentication, row-level security and embedding, hardening, backup and restore, documentation and handover. DeployOpen works on a fixed-scope basis, with scope, deliverables and timelines agreed in writing before work starts, and it does not publish a price list. Ask for the quote to list each deliverable, the environments covered, the number of embedded views and data sources in scope, what is excluded, and which items recur after go-live such as retainers for upgrades and support.
How long does a rollout take and what does it need from our team?
Timing depends on the number of data sources, the identity system that must drive access, how many embedded views you need and how much tenant isolation testing your security team requires. The work needs an accountable product owner, an identity or SSO contact who can supply group or claim mappings, a data owner for each source, a cloud or infrastructure owner for hosting, and a security reviewer for the threat model. Ask any partner for a written timeline with dependencies listed, and check which of your teams must be available in each phase.
Can our engineers build and run embedded analytics themselves?
Yes, if you have engineers who can operate a Flask or Java application, a relational metadata database, a cache and background workers, and an analytics database. The Superset upgrade documentation asks you to back up the metadata database, run superset db upgrade and superset init, test in staging and check UPDATING.md for breaking changes. Metabase runs its application database migrations automatically, so backups before each upgrade and a plan for major versions matter. The honest test is whether your team can sustain these tasks while also owning the product roadmap, so compare internal headcount against a partner quote for the same scope.
What recurring work follows go-live?
Recurring work includes platform upgrades, security patches, backup verification, capacity growth in the analytics database, identity and group mapping changes, and support for new dashboards and tenants. ClickHouse, a common analytics database, supports LTS releases for a year after release and recommends a pre-production environment with representative data before upgrades. Metabase LTS versions are supported for 14 months, and Metabase 64 is an LTS release with end of life on 1 December 2027. Ask whether the quote includes a retainer for these tasks and how restore tests are scheduled.
How do Power BI and Looker pricing models compare with open source?
Commercial BI tools price by user, capacity or both, so cost grows with audience size or data volume. Power BI Pro is licensed per user, Premium Per User raises model size to 100 GB and refreshes to 48 per day, and Fabric capacity is sold as an annual reservation or pay-as-you-go, while Power BI Embedded is quoted by sales. Looker Standard covers up to 50 internal users, Enterprise has unlimited users, and Looker Embed adds signed embedding and higher API allowances. Compare these with your projected external user count, refresh frequency, and the cost of the infrastructure and partner work that open source requires.
What recurs after go-live and how to budget for it
The largest cost that continues after go-live is keeping the platform inside vendor and project support windows. Every component in a self-hosted embedded stack has its own release train, so the budget has to cover patching and upgrades for each one, not only the first deployment. For the analytics database, the ClickHouse security policy lists versions 26.9, 26.8, 26.7 and 26.3 as currently supported with security updates, and all other versions are unsupported. Confirm which line you are on before launch and plan a recurring upgrade slot that moves you back onto a supported line before it lapses. For Metabase, LTS releases are supported for 14 months, while non-LTS releases usually reach end of life after two months, so a non-LTS choice creates upgrade work several times a year.
Upgrades carry a fixed procedure whose cost is mostly staff time and staging capacity. For Apache Superset, the upgrade documentation calls for backing up the metadata database, running superset db upgrade and superset init, testing in staging first, and reviewing UPDATING.md for breaking changes between versions. Metabase upgrades run application database migrations automatically, so take a backup of the application database first and, for major upgrades, reduce a cluster to a single node because of the migration lock. Ask any partner to price these as repeatable runbooks and to state whether staging environments are included in the retainer or billed separately.
Backups and monitoring are the recurring costs teams most often underestimate. The ClickHouse backup documentation notes that replication alone does not protect against human error or corruption, and warns that a backup never test-restored will probably not restore when needed. Budget for scheduled restore tests on a spare cluster, storage for backups in S3-compatible or Azure Blob Storage, and alerting on Keeper health, replica lag, failed Superset alerts and reports, and Metabase availability. Decide whether a retainer or an in-house on-call rota covers these tasks, and in either case confirm response expectations, out-of-hours coverage and who owns the decision to upgrade in production.
How DeployOpen scopes and prices an embedded analytics implementation
DeployOpen plans, implements, integrates, hardens, documents and supports self-hosted open-source platforms inside the client's own cloud accounts, data centers or air-gapped networks. For embedded analytics cost, that means turning the build, partner or SaaS question into a written scope that names every component, licence obligation and recurring cost before any work starts.
How we scope
- Inventory which dashboards, audiences and data sources the embedded analytics work must serve.
- Document the chosen embedding model, licence position and plan tier, with the cost components each option carries.
- Agree in writing the infrastructure, analytics database, security and support items, so the quote follows from the written scope.
How we build
- Deploy the chosen platform, its metadata database and analytics database in the client's own infrastructure.
- Configure SSO, row-level security and role mapping, then test tenant isolation with representative test accounts.
- Test query load against non-production data before any dashboard is exposed to customers or partners.
How we hand over
- Deliver architecture, configuration and licence records so the client can see every paid and unpaid component.
- Document backup, restore testing and the upgrade runbook, including the steps to back up the metadata database first.
- Hand over a cost model listing which items are one-time and which recur, so budget owners can plan renewals.
How we support
- Offer ongoing support, upgrades and continuous testing as a retainer, separate from the fixed-scope build.
- Re-check licence terms and plan limits when user counts, distribution model or product features change.
- Confirm the client's upgrade testing and security patch process with their own team and with the vendor's release notes.
