Ryz Labs/Replace/Replace Tableau
Replace

Replace Tableau with Owned Dashboards and a Warehouse: Runbook

For data leaders and CFOs facing a Tableau renewal: how mixed seat licensing adds up, what a warehouse-plus-dashboards rebuild involves, and where the cost lines cross.

Moving off Tableau is worth modeling once hundreds of employees mostly read a fixed set of dashboards, those charts belong in your own product or ops tools, and you have data engineers ready to own a warehouse and a charting layer long term. Where analysts explore freely and publish fresh workbooks weekly, Tableau usually earns its keep. Your role mix decides it: Creator, Explorer and Viewer counts, each × its price × 36 months, set against $7,000 × engineers × build months plus run cost.

How Tableau licensing adds up

Tableau charges by role, and each role's rate depends on edition. For Tableau Cloud, a Standard Creator lists at $75 per user per month, billed annually on an annual contract (list price checked November 21, 2026; check your contract). Each deployment must include a Creator. Tableau Cloud+ and the Tableau+ Bundle go through sales and carry no public price.

Edition, roleList priceUnitNotes
Standard, Creator$75User, monthlyAnnual contract; one Creator minimum
Standard, Explorer$42User, monthlyAnnual contract
Standard, Viewer$15User, monthlyAnnual contract
Enterprise, Creator$115User, monthlyAnnual contract
Enterprise, Explorer$70User, monthlyAnnual contract
Enterprise, Viewer$35User, monthlyAnnual contract
Cloud+ and Tableau+ BundleQuote onlyContact salesNo published figure

List prices checked November 21, 2026. Check your contract; negotiated prices differ.

Growth at renewal follows a recognizable shape. Viewers cost least per head yet usually outnumber everyone, so publishing one dashboard to a new department adds dozens of $15 lines in a single step. Upgrading someone from Viewer to Explorer so they can slice data on their own takes them from $15 to $42 on Standard, close to triple. Moving the site from Standard to Enterprise lifts every role: Creator $75 to $115, Explorer $42 to $70, Viewer $15 to $35. Enterprise also brings larger site storage and higher API allowances, which is often the reason for the move. Before any rebuild model, ask your admin for a count per role and for how many Viewers used a dashboard in the last quarter.

Keeping Tableau versus owning the dashboards

What replaces Tableau is seldom another BI product. More often it is the warehouse you already pay for, a metrics layer defined in version-controlled SQL, and either an open-source dashboard tool you host or charts built straight into your application, all maintained by your own engineers. That suits some data orgs and burdens others.

SituationRecommendation
A few dozen seats, mostly analysts authoring new views every weekRenew. Free-form visual exploration is the hardest piece to rebuild, and spend is modest.
Nobody available to own pipelines and data modelsRenew. Charts are only as trustworthy as the tables beneath them.
Analysts are fluent in Tableau and author their own workbooksRenew their seats; retraining an analytics group has a real price.
Hundreds of Viewers opening the same fixed operational dashboardsPrice out owning those; the Viewer count is what scales with headcount.
Customers or partners want charts in your productBuild them into your application against your own warehouse.
Metric definitions must be versioned, reviewed and identical everywhereOwn the metrics layer as code, regardless of what draws the charts.
Middle pathHold a handful of Creator and Explorer seats for exploratory analysis; rebuild the high-traffic Viewer dashboards as owned code on the same warehouse.

Of all the options, the hybrid row moves the numbers most, since it drops the fastest-growing seat type and leaves analyst habits untouched.

Seat-mix payback over three years

Price each role (Creators × $75, Explorers × $42, Viewers × $15 on Standard), sum them, and multiply the monthly total by 36 to get the licensing line. The ownership line starts with $7,000 × engineers × build months, adds the Tableau subscription for each build month and for one parallel month (old dashboards stay up until people trust the new ones), and then adds run cost from that parallel month on. Payback is the first month from which ownership's cumulative spend never rises above licensing's cumulative spend. The run figure is our placeholder, not a quote: part of a data engineer's week plus hosting for the charting layer. We treat the warehouse as already paid for, since it exists with or without Tableau on top; if yours does not exist yet, put it into run cost.

Creator / Explorer / ViewerLicensing per yearLicensing, 3 yearsBuildRun per year (placeholder)Ownership, 3 yearsPayback month
5 / 10 / 35 (50 people)12 × ($375 + $420 + $525) = $15,840$47,520$7,000 × 2 engineers × 4 months = $56,000$15,600, i.e. $1,050 (15% of an engineer) + $250 hosting each month$56,000 build, $6,600 Tableau for 5 months, $41,600 run for 32 months: $104,200Never: ownership is only $20 a month cheaper once live
15 / 60 / 225 (300 people)12 × ($1,125 + $2,520 + $3,375) = $84,240$252,720$7,000 × 3 engineers × 5 months = $105,000$48,000, i.e. $2,800 (40% of an engineer) + $1,200 hosting each month$105,000 build, $42,120 Tableau for 6 months, $124,000 run for 31 months: $271,120Month 43, after year three
40 / 200 / 1,000 (1,240 people)12 × ($3,000 + $8,400 + $15,000) = $316,800$950,400$7,000 × 4 engineers × 6 months = $168,000$120,000, i.e. $7,000 (two engineers at half time) + $3,000 hosting each month$168,000 build, $184,800 Tableau for 7 months, $300,000 run for 30 months: $652,800Month 18

Tableau Standard list prices as of November 21, 2026; check your contract. Engineering at Ryz's $7,000 starting monthly rate. Run figures are placeholders for your own.

With 50 people, Tableau costs $1,320 a month against a $1,300 placeholder for upkeep, so ownership never catches up and renewing is right. With 300, payback arrives in month 43, essentially even across a three-year term. With 1,240, it arrives in month 18, because Viewer and Explorer spend is large while owned run cost stays flat as headcount grows. Enterprise pricing would raise every licensing figure and pull each payback month earlier. The hybrid option is not modeled here; it keeps some seats and shrinks the build. Try your own role counts in the renewal versus custom build TCO calculator.

The rebuild, stage by stage

Stage 1: Map content, lineage and who looks at what

Tableau's Metadata API is one GraphQL endpoint that can track lineage between content and external assets, which lets you enumerate every workbook, the data sources it reads and the upstream tables behind those. Combine it with whatever last-quarter usage data your admins can export. Most sites carry a long tail of workbooks untouched for months; those get archived, never rebuilt. Finished when: dashboards are ranked by audience and traffic, each traced to source tables and marked rebuild, leave on Tableau, or archive.

Stage 2: Read the business rules out of workbook files

The logic lives in readable files. Per Tableau's file reference, .twb is the workbook and .twbx zips the workbook with local data and images; .tds carries a data source's connection, calculations and groups but no rows; .hyper is an extract. Calling Download Workbook over REST returns .twb as XML or the packaged .twbx, and includeExtract=False leaves the extract behind; accounts without admin rights need ExportXml permission. Parse that XML for every calculated field, parameter and filter, since those formulas are the rules your metrics layer must reproduce exactly. Finished when: each calculated field slated for rebuild is documented with its formula and a named owner who confirms it.

Stage 3: Land extracts in the warehouse

Some dashboards query live connections; others depend on extracts that exist nowhere but Tableau. With the Hyper API (Python, C++, Java) you can create, read and update .hyper files using SQL, and it also reads CSV, Parquet and Iceberg, so each extract can be read and loaded into PostgreSQL or your current warehouse. When the original source still exists, rebuild the pipeline from it and keep the extract only as a check. Finished when: row counts and key totals agree between each extract and its warehouse table.

Stage 4: Metrics as code, then the first dashboard

Write each metric once as a SQL model in git, tested for nulls, duplicates and totals that must tie to the finance close. That layer outlasts any charting choice. Next, rebuild the top-ranked dashboard, either in a self-hosted open-source dashboard tool or as charts inside your React application where staff and customers already work. Release it behind a flag to its usual readers. Finished when: for identical filters and dates the new dashboard and the Tableau original show the same figures, and its daily readers say so.

Stage 5: Who sees which rows

Translate who-can-see-what in Tableau today into roles in the new stack, signed in through your identity provider. Put row-level rules (a regional manager sees one region) in the warehouse or metrics layer instead of individual charts, so any tool querying the data obeys them. Log each dashboard view and data export. While scripts still talk to Tableau, mind its auth: a REST sign-in token is good for 240 minutes by default, Cloud personal access tokens expire after 15 days without use, and Cloud sites with MFA require a personal access token. Finished when: a test account per role sees exactly its rows, and the log proves it.

Stage 6: Refresh schedules and a parallel close

Leave Tableau dashboards running while the replacements refresh on the same cadence, and diff the headline numbers daily. Tableau Cloud limits extract refresh runtime to 120 minutes and viz requests to 600 per hour per user, so throttle any heavy crosstab (.xlsx) or custom-view CSV pulls during this phase. Finished when: one month-end close goes by with both systems in agreement and each rebuilt dashboard's audience switched.

Stage 7: Shrink seats, archive, shut down

Repoint bookmarks and intranet links, store the final .twbx for every retired workbook, and cut seats at renewal to match what moved: Viewers first, Explorers second. For a full exit, hold one Creator until the archive is complete. On Tableau Cloud, deleting a site deactivates it immediately and schedules deletion after 30 days, with reactivation possible inside that window. We found no documented retention period for when the subscription itself ends, so archive before the contract date. Finished when: retired workbooks get zero traffic for a full reporting cycle and renewal is signed at the lower count, or skipped.

Tableau exit review list

An AI pod on your team: the opening fortnight

Week one

Week two

Where the work is mechanical, our AI pod leans on coding assistants: translating Tableau calculated fields into SQL, generating a reconciliation test for each metric, scripting workbook and extract downloads. An engineer reviews each of those diffs, tests must pass, and the analyst who owns a metric approves its definition before anyone depends on it.

Roles for a BI rebuild

This is data work first. Plan on a tech lead owning the metrics layer and switch-over, one or two data engineers for extracts, pipelines and models, one to three senior full-stack engineers for the charting layer and access control, and part-time QA on reconciliation. Many teams begin with a pod of 3 to 5 senior engineers (a tech lead plus engineers), and a full pod is about seven. A narrower scope, such as moving only Viewer dashboards under the hybrid option, can begin with a tech lead plus one or two engineers. Our business intelligence and data engineering pages cover the wider work.

Ryz engineers on your team start from $7,000 per month (our published range is $7,000–$15,000 per engineer per month, mostly mid-level to senior), quoted per team. Once you say go, engineers can typically start in under two weeks. Ryz engineers work on your team, reporting to your leads. Bring seat counts by role and scope your team.

Related

FAQ

At what seat count does owning BI beat a Tableau renewal?

It turns on role mix more than total headcount. Price every role, multiply by 36, and compare with $7,000 times engineers times build months plus Tableau fees through the build and parallel month plus run cost. At Standard list prices our model shows 50 people never paying back, 300 people paying back near month 43, and 1,240 people near month 18.

Which Tableau APIs and formats get our content out?

The Metadata API inventories content and lineage. REST Download Workbook returns each workbook as .twb XML or packaged .twbx, and non-admin accounts need ExportXml permission for it. The Hyper API reads .hyper extracts with SQL so they can be loaded into a warehouse. View crosstabs export as .xlsx and custom-view data as CSV.

What sets the length of a Tableau rebuild?

How many dashboards survive the inventory, how many calculated fields they hold, whether extract sources still exist, how intricate row-level security is, and how quickly metric owners sign off. Lineage mapping and workbook parsing in week one produce a grounded estimate, and a parallel month-end close belongs in any honest plan.

How should we budget the build?

Engineers times months times monthly rate, from $7,000 per engineer, plus run cost for hosting the charting layer, refresh compute and a share of an engineer's week. Count the Tableau subscription you keep paying through the build and parallel month as part of ownership, and add warehouse cost if you are not already running one.

Can analysts keep Tableau while Viewer dashboards move?

Yes. The usual split retains a few Creator and Explorer seats for exploration and prototyping and moves fixed, heavily viewed Viewer dashboards to owned code on the same warehouse. Analysts keep the tool they know, and the seat type that grows with headcount leaves the renewal.

Questions we didn't answer? Email info@ryzlabs.com.

Thanks — your message has been sent. We’ll get back to you soon.

Something went wrong while sending your message. Please try again or email info@ryzlabs.com.

Explore Ryz Labs

Software development servicesAI development companyIndustriesStaff augmentationAI pod teamsNearshore software developmentHire engineers by roleEngineer ratesGuidesBuyer guidesCase studiesHow we vet engineers

Ryz Labs

Senior engineers on your team. AI pod teams that ship.

Tell us what you're building. You get a scoped plan, a price and the names of the people who would do the work.

  • Only the top 1% of tens of thousands interviewed make it
  • On US business hours, including New York hours
  • Trusted by Fortune 500 engineering teams

Tell us who you need

A Ryz partner replies with a scoped team plan.

Start a conversation →

Prefer to talk? Book a 15-min call →