FAQ - MyDataWork

What is MyDataWork?

MyDataWork is a workspace for data analysts and teams both small and large to organize and share the files, workflows, dashboards, notebooks, and SQL behind their data work — and connect that work to the business outcomes it supports. It brings together asset cataloging, use case tracking, lineage visualization, AI-assisted recommendations, portfolio reporting, and integrations with tools like Jira — all in one place built specifically for how analysts actually work.

Who is it for?

MyDataWork is designed for data analysts, analytics engineers, and analytics managers who want a practical way to organize their work, demonstrate its value, and hand off context to colleagues. It is particularly useful for analysts who manage recurring work across many files and tools, work across multiple business domains, or need to communicate the impact of their data work to stakeholders and leadership.

Is there a walkthrough to help me get started?

Yes. Click Help ▾ → “Take the guided tour” in the top-right of the app any time. The guided tour highlights each step on the live screen and walks you through the five things that get you to first value: connect your work → frame a use case → link the assets that feed it → see it roll up on the dashboard → try an AI action. At each step you can do the task right there in the live app or just read and move on — you stay in control — and it’s fully skippable and re-runnable. There’s also an optional “Scope an agent” tour for Agent Studio.

Is MyDataWork a data catalog?

Not in the traditional sense. Enterprise data catalogs are typically IT-managed platforms designed to inventory data sources, enforce governance policies, and serve data engineers and platform teams. MyDataWork is analyst-facing and workflow-oriented — it organizes the work analysts do with data, not the data infrastructure itself.

That said, MyDataWork is highly complementary to data catalog investments. Where a catalog answers “what data exists and where does it come from,” MyDataWork answers “what are analysts doing with that data, why does it matter, and what value is it generating.” Used together, they complete the picture from raw data asset to business outcome.

Is MyDataWork just a note-taking, knowledge management, or project tracking tool for analysts?

Not quite — though the comparisons are understandable. MyDataWork shares some surface-level characteristics with personal knowledge management tools, project trackers, and collaborative documentation platforms. But it is designed around a specific problem none of those tools were built to solve.

How it differs from personal knowledge management tools: Personal knowledge management tools are built around the idea that the individual should own and organize their own working context, which is a philosophy MyDataWork shares. But the similarity mostly ends there. The fundamental difference is that MyDataWork knows what your files actually are. Rather than asking you to describe your work in notes and build your own structure, MyDataWork reads the metadata from your actual working files — SQL scripts, Excel models, Python notebooks, Alteryx workflows, Tableau workbooks, and cloud assets — and builds the asset catalog and lineage map automatically. You do not write “I have a file that references this database table.” MyDataWork already knows.

How it differs from project trackers like Jira: Project trackers are built around tasks, tickets, sprints, and software development workflows. They are excellent at tracking what needs to be done and who is doing it. What they do not understand is the file-level context behind data work — which Excel models feed which dashboards, which SQL scripts power which reports, which Alteryx workflows support which business decisions. A Jira ticket can record that an analyst is “working on the quarterly forecast” but it cannot tell you which files are involved, what data sources they depend on, whether those dependencies are tracked or invisible, or what the measurable business outcome of that work is. MyDataWork captures all of that automatically from the files themselves. It also integrates with Jira directly — analysts can push structured use case summaries to Jira issues with professional templates and progress updates, so teams using Jira for organizational tracking do not have to choose between the two.

How it differs from collaborative documentation platforms: Documentation platforms like wikis and knowledge bases are built for writing and organizing text — pages, articles, meeting notes, process documentation. They are valuable for institutional knowledge but passive by nature: someone has to write everything, keep it current, and remember to link things together. MyDataWork is active — it detects relationships between files automatically, flags external dependencies that are not yet tracked, surfaces outliers in reported values, and uses AI to identify automation opportunities and modernization options. The context it captures is a byproduct of the analyst’s actual work, not a separate documentation effort on top of it.

What MyDataWork is actually competing with: Most analysts have no system at all — no asset catalog, no use case tracker, no outcome record, no portfolio. They manage context in spreadsheets, email threads, and institutional memory that walks out the door when someone changes roles. MyDataWork is designed for that majority: analysts who want something that works immediately, understands their toolchain, connects their work to business outcomes, and helps them communicate their value without becoming a project in itself. For teams already using Jira or documentation platforms, MyDataWork is not a replacement — it is the analyst-facing layer that those tools were never built to provide.

How does MyDataWork support IT and data platform investments?

This is one of the most important and undersold benefits of MyDataWork. Data platform teams — those who manage data warehouses, BI tools, ETL pipelines, and governance platforms — are frequently asked to justify the cost and complexity of the systems they operate. MyDataWork helps answer that question directly.

When analysts document their use cases and connect them to the files and data sources they depend on, they are implicitly documenting which data platforms, datasets, and tools are generating real business value. That information — aggregated across a team or organization — tells a compelling story: which systems are actively used, what decisions they inform, what outcomes they produce, and where investment is paying off.

For IT leadership, data governance teams, and platform owners, MyDataWork provides something they rarely have: bottom-up evidence of data utilization and business impact, built organically from the daily work of the analysts who depend on their systems. This supports budget justification, governance reporting, and strategic planning in ways that top-down inventory tools cannot.

Does it replace our existing tools?

No. MyDataWork is designed to work alongside the tools your team already uses — Alteryx, Tableau, Power BI, Python, SQL, Excel, and others. It does not replace them or require any changes to how your team works today. Analysts continue working in the tools they know; MyDataWork simply helps them organize, document, and communicate the context around that work.

What lineage does MyDataWork build automatically?

MyDataWork builds cross-tool lineage automatically in two ways.

(1) Shared warehouse tables. When an upstream asset (SQL, Python, Alteryx, Dataiku, dbt, or a warehouse like Snowflake, BigQuery, or Redshift) and a downstream asset (Power BI, Tableau, ThoughtSpot, Looker, Sigma) reference the same table identifier, an edge is drawn. A schema-qualified match (for example prod_db.sales.orders) shows as Confirmed; an unqualified table name shows as Possible.

(2) File-level matches. When an Alteryx workflow or notebook reads a local file that is also in your catalog (for example customer_demand.xlsx), MyDataWork links them automatically as a Likely edge — a filename match worth a glance to confirm.

Because of this, Excel and CSV files are no longer always manual leaf endpoints: where a workflow references one by filename, the edge is inferred for you. You can still add a manual edge anywhere a relationship isn’t present in the metadata. Every edge carries its confidence tier (Confirmed / Likely / Possible) plus a plain-language evidence string shown on hover, so the graph tells you how solid each link is. Cloud-warehouse sources like Snowflake, BigQuery, Redshift, and dbt emit fully-qualified table names, so their connections tend to be Confirmed; local files often connect by filename as Likely — that’s expected.

Same-tool pairs — two SQL files referencing the same table, for example — are treated as siblings of a shared source rather than a parent/child dependency, and surface in the related-assets card on each asset’s detail panel rather than as an edge in the graph.

What can MyDataWork connect to?

MyDataWork connects to twelve cloud sources and to local files.

Cloud sources (configured in Setup → Connect your data → Cloud Sources; one-click setup, no installation; read-only credentials):

Data warehouses

Snowflake — reads INFORMATION_SCHEMA for catalog, schema, table identity, column names and types, row counts, and last-altered timestamps.
BigQuery — reads dataset, table, and column metadata via the BigQuery metadata APIs (Metadata Viewer role; no query cost, no row access).
Redshift — reads INFORMATION_SCHEMA for catalog, schema, table, and column structure, plus best-effort row estimates.
Databricks — exports each notebook’s source and parses it for table references.

Repositories & transformation

GitHub — downloads file source from the configured repositories, parses it for tool type and table references, and retains only the structural signals.
dbt Cloud — downloads each project’s latest run manifest.json to catalog models, sources, and the dependency graph between them.

BI & analytics — catalog the dashboards, workbooks, and datasets themselves, plus their source-table lineage:

Power BI — read-only admin Scanner API: workspaces, datasets, reports, and dashboards, plus dataset→source lineage.
Tableau — Metadata API: workbooks and published datasources, plus their upstream source tables.
Sigma — workbooks and their source references.
Looker — dashboards and looks.
ThoughtSpot — liveboards and worksheets.

Data science

Dataiku — projects, datasets, and recipes, with dataset table references.

Local files via the Windows Connector. Recognized file types: Excel (.xlsx / .xlsm), CSV, SQL, Python (.py / .ipynb), Tableau (.twb / .twbx), Power BI (.pbix / .pbit), Alteryx (.yxmd / .yxwz / .yxmc), Dataiku project exports (.zip), ThoughtSpot (.tml), and Looker LookML (.lkml).

Many other tools — Hex, DataRobot, Fivetran, Matillion, Coalesce, Prophecy, Datameer, Airbyte — write their work products as tables into Snowflake, Databricks, BigQuery, or Redshift. Once the warehouse is connected, those outputs are cataloged through the warehouse connection, with no tool-specific integration needed. Tools that store their data only in their own backend (Smartsheet, Airtable, Monday, Notion, Salesforce, HubSpot, NetSuite, and most SaaS app data) need to be replicated to the warehouse first — once replicated, MyDataWork catalogs the resulting warehouse tables.

MyDataWork does not currently offer direct connectors for Google Drive, OneDrive, SharePoint, Dropbox, or other generic cloud file storage. Files kept in those services are picked up when their sync folders sit on your local disk and are scanned by the Windows Connector like any other local folder.

How are BI tools (Power BI, Tableau, Looker, ThoughtSpot, Sigma) cataloged?

BI tools can be cataloged two ways. Connect them directly (Setup → Cloud Sources) to catalog the dashboards, workbooks, and datasets themselves, plus the source tables they read — that’s the richest picture. Or rely on your warehouse connection, which catalogs the source-of-truth tables those dashboards read (a shared, schema-qualified table shows as Confirmed lineage; an unqualified name as Possible). The warehouse route shows the underlying tables; the direct connector adds the dashboards and their lineage.

Can it help with onboarding new team members?

Yes — significantly. One of the most common pain points for data teams is knowledge transfer. When an analyst leaves or a project changes hands, the context behind recurring work — which files matter, why they exist, what decisions they support, what to watch out for — is often lost. MyDataWork captures that context as a natural byproduct of daily work, making onboarding faster and reducing dependence on institutional memory held by individuals.

What kinds of assets can MyDataWork organize?

MyDataWork supports any file-based data asset your analysts work with, including Excel and CSV files, SQL scripts, Python notebooks, Alteryx workflows and macros, Tableau and Power BI workbooks, ThoughtSpot TML files, Dataiku DSS project exports, Looker LookML files (.view.lkml, .model.lkml, .dashboard.lkml), and supporting documentation. Assets are connected to the use cases they support, the people who own them, and the lineage that shows how they relate to each other.

Can I connect my own files?

Yes. The MyDataWork Connector is a lightweight Windows application that indexes the files on your computer or network drive. It parses your files on your own machine to capture structure — file names, paths, sizes, creation and modification dates, plus tool type and schema-level references such as table, column, and sheet identifiers — and syncs only that structural metadata to your workspace. Your file contents and data values never leave your computer. The Connector is digitally signed (code-signed) by MyDataWork, LLC through a public certificate authority (Azure Trusted Signing), so Windows verifies the publisher’s identity when you install it. (Cloud sources work differently — see “Is my data safe?” below for what each cloud connector reads and retains.) You can then add context, tags, use cases, and relationships on top of your real working files.

Is my data safe?

Only structural metadata is retained. The local Windows Connector parses your files on your own machine to extract structure — file names, paths, sizes, modification dates, and schema-level references such as table and column identifiers — and sends only that metadata to your workspace; your file contents and data values never leave your computer. Cloud sources do read file source content or warehouse/BI metadata in order to extract structure — for example GitHub file content, Databricks notebook source, dbt manifests, warehouse INFORMATION_SCHEMA (Snowflake, BigQuery, Redshift), and read-only BI/analytics metadata APIs (Power BI, Tableau, Sigma, Looker, ThoughtSpot, Dataiku) — but the fetched content is parsed in the moment and discarded; only the resulting structural metadata (system, topics, table identifiers, column identifiers, dependency references) is persisted. No business data values, no table rows, no query results, and no formula or cell contents are ever read or stored. See our Security & IT page for the per-connector breakdown.

How does MyDataWork handle personal information?

MyDataWork organizes the structure of your work, not the data inside it — so the personal information it holds is limited and purposeful.

What it holds: your account details (name, email, and a securely hashed password), the names and emails of any team members an admin adds, and any stakeholder names or emails you choose to record against a use case. These are details you enter on purpose to make the workspace useful.

What it never touches: the personal or sensitive data living inside your files and tables. Because every connector reads structural metadata only — never file contents, cell values, or query results — the customer records, employee data, or other PII inside your actual data are never read, transmitted, or stored. (See “Is my data safe?”)

How it’s protected: passwords are hashed with bcrypt; card numbers are never stored (Stripe handles payments); data is encrypted at rest and in transit; and where we need to remember something tied to your email — like your one free assessment — we store a one-way hash, never the address itself.

AI features: the stakeholder names and emails you record are stripped before anything is sent to our AI provider — the AI sees the shape of your work, not your contacts. The only personal data that can reach it is whatever you type into free-text fields like descriptions and notes, which is entirely in your control. File contents are never sent.

You can export everything (Setup → Data portability) or delete your account and all data at any time. We don’t sell data or use it for advertising. Data Processing Agreements are available — contact contact@mydatawork.com.

How does smart file filtering work?

MyDataWork automatically filters files to help you focus on current work while keeping everything accessible. The system provides four filter options: Current (30 days) is the default view showing files modified or accessed recently — your active work; Recent (30-90 days) displays files from 1-3 months ago that aren’t stale but not current; All files provides complete catalog access with no filtering; and Archived (90+ days) contains historical files not recently updated. This intelligent categorization ensures you see the most relevant files first while maintaining access to your complete asset history.

How does progress tracking work?

In the Objectives & Progress tab of any use case, you set three values: a baseline (where things stood before the work began), a current value (where things stand now), and a target (the goal you are working toward). MyDataWork calculates progress automatically and displays a progress bar — no formula needed.

The baseline is the most important value to set correctly, and it can only be truly accurate once — at the very beginning, before the work changes anything. Record it as soon as you create a use case. The current value is updated as work progresses, and the progress bar recalculates automatically each time.

Progress is expressed using professional value measurement templates that go beyond monetary estimates: Time Savings (hours per week, days to completion), Quality Improvement (error rates, accuracy scores), Efficiency Gains (automation percentage, throughput), Risk Reduction (incident rates, compliance scores), and Cost Efficiency (cost per unit, resource utilization). Each template provides recommended units of measure, best practice guidance for baselines and targets, realistic improvement ranges, and industry-standard measurement approaches.

The result is a live, honest record of where a deliverable stands — ready to share at any moment without reconstruction or estimation.

How does use case lifecycle management work?

MyDataWork provides complete project lifecycle management with three distinct actions: Archive provides soft delete functionality that removes use cases from active view while preserving all data and history; Restore brings archived use cases back to active status; and Delete offers permanent removal for completed or abandoned projects. Users can switch between Active and Archived tabs in the Use Cases section to manage the full lifecycle. This system is particularly useful for completed projects you want to keep for reference, seasonal use cases that aren’t currently relevant, or maintaining a clean active workspace while preserving institutional knowledge.

Where do I see workspace health at a glance?

The Dashboard tab is the first tab in every workspace and surfaces six panels of workspace health designed to be read in seconds:

Use case staleness — distribution of your active use cases across freshness buckets (under 30 days / 30–60 / 60–90 / 90+).
Portfolio totals — estimated vs realized value across active use cases, plus the realization rate. Use cases with no linked assets are excluded from this total (and from the value chart) so the headline number reflects measured work; the subtext “Includes N of M active use cases” appears when any are excluded.
Stakeholder coverage — how many of your active use cases have an assigned stakeholder.
Value distribution — realized values plotted against the workspace mean and median (the robust-statistics check that also powers the Portfolio outlier hint).
Tool distribution — asset count by tool family.
Workspace Agent activity — findings surfaced and acted on in the last 14 days.

The Dashboard reflects what you see on the Assets tab — your own assets plus any shared to the team — so the numbers always reconcile with the rest of the app. Admins who need to see the full workspace catalog (across all team members’ private assets) use the Workspace view on the Assets, Use Cases, and Workspace Agent tabs.

When demo data is loaded, a small demo / real toggle in the upper-right lets you flip between the demo view and your real-data view. The toggle is dashboard-only — it doesn’t change what’s shown in Assets, Use Cases, Workspace Agent, or the Assessment.

Can I see and open all my work from one place?

Yes — the Workspace view (Dashboard → Workspace sub-tab) is a home base for your cataloged work. It lists every asset you can see, grouped by tool (Excel, Power BI, SQL, Tableau, Python, cloud sources, …) and ordered alphabetically, showing how recently each changed and how many use cases it belongs to. Local files open on your own machine in their default app (via the Windows Connector); cloud assets open in-app. It’s a convenience view over your existing catalog — it doesn’t change what’s cataloged, your data, or who can see what (your private-by-default assets stay private). The Dashboard’s default view stays the at-a-glance Overview; Workspace is one click away.

What is the Asset Estate Assessment?

An AI-powered review of your workspace, runnable on demand from Setup → Asset Estate Assessment. The first run is free on every plan — one free assessment per email, forever. You can re-run the assessment any time afterward; subsequent runs use AI credits like other AI features. The free-first-run affordance is keyed by a SHA-256 hash of your email address (the plaintext email is not stored in the eligibility log), so if you’ve used your free run in one workspace, a second one in a new workspace under the same email is routed to the paid path.

What it produces: a five-section report — Estate overview, Connections worth making explicit, External dependencies, Use case opportunities, and Health observations — plus a Concrete next actions list ranking the top items. Exportable to PDF, with the workspace name, who it was generated for, and the timestamp on the export.

What it reads: asset metadata (names, paths, tool types, topic tags, lineage edges) and use case text (titles, objectives, descriptions, value figures, stakeholder roles) only. File contents are never read — the same boundary that applies to every other AI surface in the app.

When to run it: after you’ve connected real assets and configured at least 2 active use cases (the assessment is use-case-driven; with fewer than 2 active use cases the Run button is disabled). If you run with fewer than 10 real assets you’ll see a soft-floor confirmation; you can still proceed, but thin estates produce thin output.

Demo path: preview the report format any time without using your free run — the View demo assessment link renders a curated example using the same template the real run produces.

Workspace-grounding filter: every finding names a specific asset, use case, or stakeholder from your workspace. Generic template phrasings (“consolidate your KPI dashboards”, “merge your forecasting models”) are dropped during synthesis if no workspace-specific evidence backs them — so the assessment reads like a senior analyst who has audited your estate, not a checklist of generic advice.

How does the Workspace Agent compare to the Asset Estate Assessment?

Both are AI-powered, and both can be scoped to your whole workspace or to a specific client/project group — but they do different jobs, and the difference is in what they produce.

The Workspace Agent is a detector. It runs a set of rule-based checks across your use cases and assets and surfaces discrete, specific findings worth acting on — reuse opportunities, automation candidates, missing stakeholders, gaps. Think of it as a running worklist: run it any time for an updated set of individual items to address. (The checks are rule-based; AI is used only to phrase each finding in plain language — so it is accurate and grounded, never a black box.)

The Asset Estate Assessment is an analyst. It produces a point-in-time synthesis — a connected, narrative review of your whole estate’s health and readiness that pulls the Agent’s findings together with its own analysis into a comprehensive report. Run it periodically for the big picture, not as a running list.

In short: run the Workspace Agent for an ongoing, specific to-do list; run the Estate Assessment for a periodic, comprehensive read on where the work stands.

How does MyDataWork help me ensure my reported values are accurate?

MyDataWork includes a sophisticated data quality layer that watches for value measurements that look statistically unusual relative to your other work. Once you have at least 4 use cases with non-zero values entered, if any estimated or realized value is significantly higher or lower than the rest of your workspace, a gentle amber prompt appears asking you to confirm the figure before you share it with stakeholders.

The Portfolio tab also compares the mean and median realized value across your use cases — once you have at least 3 use cases with realized values, when these figures diverge significantly it signals that one use case may be disproportionately affecting your portfolio total.

These checks are informational only and never block you from using the app. They are designed to help you walk into stakeholder conversations with numbers you are confident in, not numbers that were accidentally inflated or understated.

This approach draws on the principles of robust statistics, a field concerned with producing reliable estimates even when data contains outliers or errors.

Can MyDataWork show data sources that my assets depend on but that aren’t in my catalog?

Yes. In the Assets tab, any data source detected in a file that does not match a cataloged asset is flagged with an amber “Not in catalog” badge next to the data source name. This makes invisible dependencies visible — for example if a SQL script references a database table that is not tracked in your workspace, you will see it called out immediately.

The Lineage tab surfaces these as ghost nodes — dimmed, dashed nodes — automatically. The Assets-tab lineage preview has a “Show external dependencies” toggle to show or hide them in that smaller view.

When exporting your portfolio, you can optionally include external dependencies in the lineage diagrams by checking “Include external dependencies in lineage” in the export options. This is useful when sharing with IT teams or stakeholders who need to understand the full dependency picture.

What are Solution Packs?

Guided head-starts for a new workspace. In Setup → Solution Packs, pick a scenario and one or more topic areas, and MyDataWork adds a curated set of best-practice starter use cases — editable examples you tailor, not AI-generated — grouped into an Initiative named for the pack, plus an empty asset group ready to hold your own files. Six packs are available today: FP&A Modernization, Power BI Portfolio Review, Agentic Analytics Readiness, Consulting Discovery Accelerator (run per-client by consultants), Alteryx Modernization Readiness, and Analytics Value Realization Review. Solution Packs are free on every plan (including the Explorer trial), use no AI credits, and re-running a pack never duplicates what you already have.

What is an Initiative?

A project or program that groups related use cases — the use-case counterpart of an asset group (which groups assets). Initiatives are personal to you, and each use case belongs to one initiative, so dashboard value roll-ups never double-count. Manage them on the Use Cases tab’s By initiative view, and filter the Dashboard, Portfolio, and Insights to a single initiative.

Can I organize my assets and use cases by client or project?

Yes, in two complementary ways. Asset groups organize your files: create a group (e.g. one per client), assign assets to it (an asset can belong to more than one group), and scope any AI analysis to that group — the Workspace Agent, Estate Assessment, use case proposals, recommendations, and the Leverage modes all run against just that client or project for focused, per-engagement results. Scoping never changes the credit cost. Initiatives organize your use cases: group related use cases into a project or program to organize the Use Cases tab and roll up value by initiative on the Dashboard, Portfolio, and Insights. Every AI analysis can be scoped to a group or an initiative. This is especially useful for consultants and teams who manage work across several clients or domains in one workspace.

Do I have to turn AI on before I can use the AI features?

No. AI is enabled by default for new workspaces. An admin can turn it on or off any time in Setup → AI Assist. AI only runs when you explicitly ask for it (for example, proposing use cases, a Leverage analysis, the Assistant), and each AI action draws from your plan’s daily AI-credit allowance — Explorer 3/day, Solo 5/day, Team 20/day per user, with the option to buy more.

How do AI credits work?

MyDataWork uses a daily credit system for AI-powered features, and the daily allowance is per user: Explorer 3 credits per day (with a 60-credit total cap across the 90-day trial); Solo 5 credits per day; Team plans 20 credits per day for each member — not a shared daily pool.

AI features that consume credits include use case proposals (1 credit each), AI recommendations and analysis, leverage analysis modes, the MyDataWork Assistant, and the Workspace Agent (Analyze). The Workspace Agent draws from the same daily AI credit pool as other AI features once its free attempts are used — 1 credit per finding it phrases, with a ceiling of 8 credits per Analyze run (reached only if all six rules find candidates). The Workspace Agent gives you up to two free attempts that don’t use your daily AI credits: the first run that surfaces findings is your free run, and if a run finds nothing you get a second free try so an empty first scan doesn’t cost you. See “What does the Workspace Agent do?” below.

You can purchase additional credits if needed: 10 credits for $5, 25 credits for $10, or 50 credits for $15. Purchased credits go into a shared workspace pool that any member draws from automatically once their own daily allowance is used up, and only a workspace admin can purchase credits for the pool (the account owner is an admin). Your daily credits reset each day and don’t roll over; purchased credits carry over until used and never expire. Your first use case proposal is free as a one-time trial. Solution Packs and Initiatives cost no credits.

What AI-powered features does MyDataWork include?

MyDataWork includes comprehensive AI-powered features that consume daily credits:

Propose Use Cases (1 credit each) analyzes your workspace and asset patterns to generate specific, actionable use case proposals based on your actual files and workflows.

Leverage Analysis Modes: Find reuse opportunities identifies where your use cases overlap and where work could be consolidated across projects or team members. Identify automation candidates analyzes your use cases and assets to find business processes that could be candidates for intelligent automation, with estimated value, complexity ratings, and suggested next steps. Discover marketplace data recommends external datasets available on your configured cloud platforms (AWS, Google Cloud, Snowflake, Azure, Databricks) that could enrich your use cases. Migration Assist analyzes individual assets or complete use case workflows to identify opportunities to migrate or modernize tools, with effort estimates, expected benefits, confidence ratings, and concrete next steps.

AI Recommendations analyze your use case descriptions and linked assets to surface improvement ideas grounded in your actual work context.

Workspace Agent is a proactive observer that scans the workspace on demand and surfaces six categories of state worth your attention. It draws from the same daily AI credit pool as the other AI features once its free attempts are used (1 credit per finding phrased, ceiling 8 per run; up to two free attempts to start). It surfaces — missing stakeholders, recently-added assets not linked to a use case, removed assets still in use, stale use cases, high-value use cases tracking below estimate, and assets serving many use cases. See “What does the Workspace Agent do?” below for details.

MyDataWork Assistant is a context-aware chat assistant available on every screen via the teal button in the bottom-right corner. It understands your workspace and current location in the app, providing specific guidance rather than generic help.

What does the Workspace Agent do?

The Workspace Agent runs on demand from the Workspace Agent tab. Click “Analyze” and the agent checks six things across your workspace and surfaces what’s worth your attention, grouped into three categories.

Cleanup (patient, not urgent): use cases that have assets and progress but no stakeholder assigned, and active use cases that haven’t been updated in 90+ days.

Activity (something changed): recently-added assets (last 14 days) that aren’t linked to any use case yet — with a suggested best-fit use case based on keyword overlap. Also: assets you removed within the last 30 days that are still referenced by an active use case or a lineage edge, so you can spot broken dependencies that followed from the removal.

Insight (strategic observation): active use cases with high projected value and low realized value (less than 25% of estimate) that haven’t been updated in 60+ days, and assets that are linked to three or more active use cases (the “hidden infrastructure” of your workspace where governance attention is warranted).

Each finding includes a one-click deep-link to the relevant asset or use case. You can dismiss with a reason (not relevant, already handled, revisit later), give feedback with a thumbs up or down, and the agent auto-resolves a finding when its condition no longer applies — for example when you assign the stakeholder, link the asset, or restore the removed file.

The Workspace Agent’s checks are rule-based; AI is used only to phrase each finding in natural language, so the findings are grounded in your actual workspace rather than generated. You get up to two free attempts that don’t use your daily AI credits — the first run that surfaces findings is your free run, and if a run finds nothing you get a second free try so an empty first scan doesn’t cost you. After that, subsequent runs draw from your daily AI credit pool — 1 credit per finding the agent phrases in natural language. A single run is capped at 8 credits, the ceiling reached only if all six rules find candidates at once; in practice most runs consume far fewer credits because most rules don’t find candidates. The exact cost ceiling for the next run is shown above the Analyze button before you click.

Can I link external notes (Google Doc, Notion, Confluence) to a use case?

Yes. Each use case has a Notes URL field on the Overview tab. Paste a Google Doc, Notion page, Confluence URL, or any HTTPS link where you keep notes for this use case. When set, an “Open notes” button appears in the panel and a small indicator shows next to the use case in the list view, so you can see at a glance which use cases have external notes attached.

Can I connect my own AI agent to my MyDataWork workspace?

Yes. Agent access (in Setup → Agent access) lets you give an AI agent you choose read-only access to the structure of your data work — your use cases, their linked assets, and the lineage between them — over the Model Context Protocol (MCP), an open industry standard for connecting AI agents to tools (for example Claude Code, your team’s own agent, or another MCP-capable tool).

You stay in control. You create a key for one specific agent, choose exactly which use cases it can read, and preview precisely what that exposes before anything is shared. The agent receives metadata only — asset names, tool types, structural schema, lineage, and use-case outcomes — never your data values or file contents, and never stakeholder names. Each asset includes a clean, structured list of the sources it reads (system, schema, and table), so your agent knows where the data lives. MyDataWork provides that map, never the credentials or the data itself — your agent reads any underlying data through its own connections.

You choose an access mode per key. Live mode lets the agent connect and read your current context on demand (available on paid plans, with a daily read cap); it works with MCP clients that accept a bearer key, such as Claude Code or a custom agent. Claude Desktop’s connector authenticates via OAuth rather than a bearer key, so it cannot use Live mode — for Claude Desktop, use Export. Export mode lets you download a point-in-time context file and hand it to your agent yourself (the only mode on the free Explorer plan, with a small daily export cap), and it works with any agent, including Claude Desktop and other chat apps. These caps are separate from your AI credits. Every grant change and every agent read is recorded in your audit log, and you can revoke a key at any time.

Does connecting an agent use up my AI credits?

No. Agent reads (in Live mode) and exports draw on their own dedicated daily allowance — the Live reads and Exports caps shown in the Agent access panel — not your AI credits. The AI credit pool that powers the chatbot, Leverage analyses, and recommendations is never touched by agent activity. The agent caps are a daily anti-abuse ceiling that resets each day and is completely independent of your AI credits: an agent reading your context on demand can only reach its own separate daily cap, it can never exhaust the AI credits you use for the chatbot or Leverage.

What is Agent Studio?

Agent Studio is where you scope an agentic use case before anyone builds it. You catalog the MCP sources your organization exposes, pick the tools an agent could use, arrange them on a simple canvas (trigger, inputs, agent, actions, outcome), and let AI help draft the surrounding definition — the value case, the agent persona, the reasoning approach, the guardrails, and the human-in-the-loop checkpoints. The result is a shareable definition you can export as a PDF or PowerPoint and hand to whoever will build and govern the agent.

It deliberately stops short of building or running anything. Agent Studio does not deploy agents, does not call tools, and does not read your business data — when it catalogs an MCP source it reads only what that server advertises: tool names, descriptions, and input-parameter shapes. Think of it as the architect’s brief, not the building.

It also includes an anti-“agent-washing” check. Not every automation needs an agent: if the work is stable, rules-based, and deterministic, Agent Studio says so rather than forcing an agent where a script or workflow would be the better answer.

How is Agent Studio different from Agent access?

They point in opposite directions. Agent access (Setup → Agent access) lets an AI agent you choose read the structure of your data work — your use cases, their linked assets, and the lineage between them — over MCP. Agent Studio is the reverse: it catalogs the tools your systems expose over MCP, so you can design an agentic use case around them.

Both are governed by the same metadata-only boundary: neither reads your data values or file contents. And because Agent Studio only ever reads what a server advertises — it never calls a tool or reads a resource — cataloging a source cannot cross into execution. Agent Studio sits upstream of runtime: you scope the use case here, and your team routes the actual tool access through their own governed MCP control point when they build it.

Can MyDataWork help me identify automation opportunities in my data work?

Yes. The Leverage tab includes an “Identify automation candidates” mode that uses AI to analyze your use cases and assets and surface specific processes that could be candidates for intelligent automation — where an AI agent could potentially handle steps you currently do manually.

Each opportunity includes a description of what could be automated, which of your assets are involved, an estimated value and complexity, and a concrete suggested next step. You can share any opportunity via email to a colleague or stakeholder directly from the app. All identified opportunities are saved in the app so you can revisit, review, and share them at any time without regenerating them.

Can MyDataWork recommend external data sources to improve my work?

Yes. The Leverage tab includes a “Discover marketplace data” mode that uses AI to recommend external datasets available through your cloud providers that could enrich or improve your use cases.

To use this feature, first go to Setup → Cloud Providers and select the cloud platforms your organization uses (AWS, Google Cloud, Snowflake, Microsoft Azure, Databricks). MyDataWork will then tailor its recommendations to datasets available on those specific platforms.

Recommendations include the dataset name, which platform it is available on, why it is relevant to your specific use cases, and where to find it in the relevant marketplace. Recommendations are saved so you can review them across sessions and dismiss ones that are not relevant.

Can MyDataWork help me evaluate whether to migrate or modernize my data tools?

Yes. The Leverage tab includes a Migration Assist mode that uses AI to analyze your assets or use cases and identify opportunities to migrate or modernize tools. For example, it might recommend moving a complex Excel model to Python, replacing an Alteryx workflow with dbt, or converting ad-hoc SQL scripts into structured dbt models.

You can analyze individual assets — selecting files grouped by tool type — or analyze a full use case to evaluate all its linked assets together as an end-to-end workflow. This is useful when you want to assess whether an entire pipeline could be modernized rather than evaluating files in isolation.

Each recommendation includes the current tool, the recommended alternative, migration effort (Low, Medium, or High), estimated benefit, key risks, and a concrete suggested next step. A confidence rating is included with each recommendation — lower confidence results include an explanatory note, for example when recommending a newer or less widely documented tool. Results are saved so you can revisit and share them via email at any time.

How do I export and import my workspace?

Go to Setup → Data portability. Export downloads a JSON file containing your complete workspace: all assets, lineage, stakeholders, use cases (with objectives, baseline/current/target values, progress notes, priorities, effort estimates, target dates), communication logs, saved AI recommendations, sharing and copy details, and action plans. Import merges a JSON file into your current workspace — existing items are updated by matching on path (assets) or title (use cases), new items are added. Nothing is deleted during import. Plan asset limits are enforced on import: if the import would push you over your plan’s cap, it is rejected before any changes are made. Lineage edges are imported directly and do not require a manual rebuild.

How does MyDataWork show who owns each asset in a team workspace?

In Team plan workspaces, each member has their own personal workspace where their assets stay private by default. When a member shares an asset to the team, it appears in the team’s Shared tab with attribution showing who shared it. Each asset’s detail panel also shows an “Added by” field indicating which team member originally connected that asset — whether via the Windows Connector or a cloud source.

For assets connected via any cloud source, the connecting user is always recorded. For assets connected via the Windows Connector, the member who installed the Connector and imported the files is recorded as the asset owner — each member’s imports are attributed to that member, and stay private to them until they choose to share.

How do Team plan workspaces handle sharing and collaboration?

MyDataWork Team workspaces use a shared bulletin board model. Each team member has their own personal workspace where their assets are private by default. When a member is ready to share an asset with the team, they click “Share to Team” on the asset’s detail panel — it appears in the team’s Shared tab where other members can browse it. If a colleague finds it useful, they can click “Copy to mine” to bring an independent copy into their own workspace.

Each member’s workspace stays their own. Use cases, initiatives, lineage, stakeholders, AI recommendations, and Insights all remain personal — only assets are shareable. The person who shared an asset (or the team admin) can unshare it at any time; copies others have already made are independent and unaffected.

This model lets team members connect their full working folders without worrying about exposing drafts or work-in-progress, while still making it easy to surface and reuse the work that’s ready for the team.

How do Team plan admins manage their workspace members?

Team plan workspaces have a single admin (the Owner) who manages the team. Admins can view all members, add new members, remove members, and transfer admin ownership — all from the Workspace management view in Setup.

Adding members: The admin creates member accounts directly by entering the new member’s name, email, and a starter password they will share with the member. There is no email invite flow. After receiving the credentials, members log in and can change their password from the Account page. Seat limits apply based on plan (2-5 for Team Starter, 6-10 for Team Growth, with the admin counting as one seat).

Transferring admin ownership: Admins can transfer ownership to any current member using the Transfer admin action on that member’s row. After transfer, the previous admin becomes a regular member in the workspace.

Removing members: Admins can remove any member from the workspace. The removed member’s assets, use cases, and other content remain with the workspace — only the member’s access is revoked.

How do Team plan admins see what their team is doing without surveilling individuals?

Team admins have access to a Team Metrics panel in the Workspace management section of Setup. The panel shows aggregate adoption metrics across the team:

Summary stats: total assets across the team, how many have been shared to the team bulletin board, total copies made, sharing rate (shared assets ÷ total assets), and reuse multiplier (average copies per shared asset)
Most-reused shared assets: the top shared assets by copy count, with attribution to the member who shared them
30-day activity timeline: shares and copies per day so admins can see momentum over time

Per-member breakdowns are intentionally not provided. The Team Metrics panel gives admins overall adoption visibility without surfacing individual contributor behavior. Members keep their autonomy; admins get the adoption story.

What happens if our Team admin leaves the company?

If the admin of a Team workspace leaves without first transferring admin to another member, the workspace can become orphaned with no one able to manage members or billing. If this happens, email contact@mydatawork.com and we can reassign admin access to another member. To avoid this situation, admins should transfer ownership before they leave or are offboarded.

Can MyDataWork push updates to Jira?

Yes. MyDataWork includes a comprehensive Jira integration that lets you create structured tickets and push progress updates directly from your use cases. The integration is available on all paid plans (Solo and Team). It includes professional templates for different project types: automation projects, reporting enhancements, and data quality initiatives. Each template creates structured tickets with sections for objectives, stakeholders, linked assets, acceptance criteria, and a direct link back to your MyDataWork workspace.

You can also push progress updates to existing Jira tickets with current measurements, value tracking, and latest progress notes. The system can read current ticket status, assignee, and priority directly in MyDataWork for complete workflow visibility.

To set it up, go to Setup → Integrations and enter your Jira instance URL and API token. For Jira Cloud use your email and an API token generated at id.atlassian.com/manage-profile/security/api-tokens. For Jira Data Center use a Personal Access Token.

The integration is one-way — MyDataWork sends to Jira, Jira does not write back. Your MyDataWork workspace remains the source of truth for your data work.

Can I use MyDataWork on my phone or tablet?

Yes, MyDataWork works fully in mobile web browsers (Safari, Chrome, etc.) for viewing and managing your analytical assets. You can view your analytical portfolio and executive dashboard, browse discovered assets and their details, check use case progress and metrics, export portfolio summaries, review asset lineage and connections, and purchase AI credits and manage account settings.

Desktop-only features: Initial connector setup (requires Windows Connector), asset discovery scans (automated scans run from desktop), and file system browsing and selection.

How do I set up connectors on mobile?

Connector setup requires our Windows Connector software, which only runs on Windows PCs. Once configured on your desktop, you can view and manage all discovered assets from any mobile device.

Can I discover new assets from my phone?

Asset discovery requires access to your local files and network drives, which is only available through our desktop Windows Connector. However, once assets are discovered, you can view and manage them from any device.

What’s the best mobile experience?

For optimal mobile experience, use landscape mode on tablets for viewing detailed asset information and lineage diagrams. Portrait mode works well for portfolio overviews and use case tracking.

How is MyDataWork priced?

MyDataWork offers a free Explorer plan for individuals plus paid Solo and Team plans:

Explorer (free, 90 days): 1 seat, 30 assets, 1 cloud source connection, 3 AI credits/day with a 60-credit total cap across the trial. No credit card required. Default for new self-service signups.
Solo Monthly: $20/month — 1 seat, 300 assets, 5 AI credits/day, individual workspace
Solo Annual: $192/year — 1 seat, 300 assets, 5 AI credits/day, save vs monthly
Team Starter: $900/year — 2-5 seats, 2,000 assets, 20 AI credits/day per member, team collaboration (private by default)
Team Growth: $1,440/year — 6-10 seats, 2,000 assets, 20 AI credits/day per member, advanced team features

All paid plans include a 14-day free trial with no charge until the trial ends. The Explorer plan is a separate free 90-day offering — see the next entry for the distinction. AI credits can be purchased separately as needed: 10 credits/$5, 25 credits/$10, 50 credits/$15.

Do you offer plans for larger organizations?

Yes. For larger teams we offer negotiated Enterprise plans with custom seat counts and higher asset, AI-credit, and agent-access limits, billed by invoice (no credit card required). Contact contact@mydatawork.com to discuss your needs.

What’s the difference between the Explorer plan and the 14-day free trial?

They’re two distinct offerings:

Explorer plan is our 90-day free workspace for individuals. No credit card required. 1 seat, up to 30 user-added assets, 1 cloud source connection, 3 AI credits per day with a 60-credit total cap across the 90-day trial. Designed for an individual practitioner to build their working practice over time. New self-service signups are placed on Explorer by default.
14-day free trial applies to our paid Solo and Team plans. It’s the standard “try before you’re billed” trial for the paid tiers. After the 14 days, billing begins on the chosen plan. You can upgrade from Explorer to a paid plan at any time during your Explorer trial or the 30-day frozen period that follows.

How do I get access?

You can register directly at app.mydatawork.com and select a plan. New signups start on the free 90-day Explorer plan by default; paid Solo and Team plans include a 14-day free trial, so you can explore all features before your trial period ends.

What’s the difference between canceling my subscription and deleting my account?

These are two different actions with very different outcomes — it’s important to choose the right one.

Canceling your subscription stops future charges and ends your paid access. Your workspace data is retained for 30 days so you can resubscribe anytime during that period and pick up where you left off. Cancel via Account → View plans & billing → Manage billing to access your Stripe customer portal.

Deleting your account is immediate and permanent. Your account, workspace, and all associated data (assets, use cases, stakeholders, progress history) are removed right away. There is no retention period and no recovery option. Delete your account from Account → Delete my account.

If you simply want to stop paying, cancel your subscription. Delete your account only if you want to permanently remove yourself and your data from MyDataWork.

How do I delete my account?

You can delete your account yourself from the Account page. At the bottom of the Account section, click “Delete my account.” A confirmation dialog explains the consequences — deletion is immediate and permanent, your data is not retained, and your subscription (if any) is automatically canceled. Complete the confirmation checkbox and click Delete my account.

Important requirements:

Super-admin accounts cannot be deleted via self-service.
If you are a Team admin with other members in your workspace, you cannot delete your account until you transfer admin ownership to another member or remove all other members. This prevents workspaces from being orphaned.
Active annual subscriptions are non-refundable — deleting mid-cycle forfeits the remaining paid time.

You are also invited to share optional feedback about why you’re leaving. This helps us improve the product.

Can I get a refund on my annual plan?

Annual plans are non-refundable. If you cancel an annual subscription, you retain access until the end of the current annual period, then lose access (with 30-day data retention). If you delete your account mid-annual-cycle, the subscription is immediately canceled and the remaining prepaid time is forfeited with no refund.

If you are on a monthly plan, you can cancel anytime; you’ll retain access through the end of the current billing period, and no future charges will be made.

What happens if I hit the AI credit limit?

AI credits reset daily and are per user — Explorer gets 3 credits per day (with a 60-credit cap across the 90-day trial), Solo gets 5 credits per day, and Team plans get 20 credits per day for each member. If you run out, purchased credits (10 for $5, 25 for $10, or 50 for $15) are drawn from automatically — they sit in a shared workspace pool that any member can use once their daily allowance is exhausted; only workspace admins can purchase them. Purchased credits are available immediately and carry over until used. You can also wait for your daily credits to reset the following day.

What happens if I hit the lineage rebuild or export limits?

Some features have daily or hourly rate limits to protect platform resources: lineage rebuilds are limited to 5 per day, Jira integration calls are limited to 20 per hour (shared across push and progress updates), and exports are limited to 10 per day. If you hit one of these limits, you’ll see a rate limit message and can retry after the limit resets. Normal workflows rarely hit these limits — they exist primarily to prevent accidental runaway usage.

Frequently Asked Questions