Key concepts and the mental model#
Most Reportworq concepts fall into place once one idea clicks. That idea is the single most important thing a new user has to get.
Three things you build#
You create everything from the New menu on the Workspace, and three of its items are the building blocks of the product:
- A distribution job (New Distribution Job) distributes reports. It takes one or more reports, such as an Excel workbook, a web report, or a PowerPoint deck, refreshes them with current data, personalizes them, and delivers the results. You build it in the Distribution Job Editor. Once a page has introduced it, these guides call it a job.
- A web report (New Web Report) is a report you build in the browser, in the report editor, rather than in Excel. Authors and administrators open it in the viewer, and a job can distribute it like any other report. Readers never see a web report itself: they receive the output of the jobs that distribute it. Web reports and data models are early release features. See How reporting works and Build a web report.
- A contribution campaign (New Contribution Campaign) runs the other direction: it collects input from the people who own the numbers and writes it back to the planning system. These guides call it a campaign for short.
The distinction that matters most is between the job and the reports inside it. A report is content: the workbook, web report, or deck you add on the job's Reports step. The job is what refreshes that content and sends it out. What a run of the job delivers, the PDF, the workbook, the email, is its output.
Limits of a web report#
A web report has three limits, and none of them cuts your data without telling you. The numbers are the first version and will be raised as testing improves.
| Limit | The number | What happens at it |
|---|---|---|
| Rows read from the source | 1,000,000 rows per list or crosstab, and no more than 2,000,000 cells per report (an administrator can raise that to 5,000,000; in the Excel analyst pane on Mac and the web, all the reports in one workbook together may also hold at most 5,000,000 cells) | A plain list with no limit of its own shows its first 10,000 rows (or the rows that fit on its sheet, if fewer) and says there are more; Load all rows reads up to the limit. A list that sorts, filters or ranks, and every crosstab on a data model, needs every row to be right, so above the limit it is not shown, and it tells you why. To read fewer rows, use a data model variable or a filter on the view in the data model (a filter you add to the report itself applies after the rows are read, so it does not help), or ask the report's author to set Rows read from the source to 1,000,000 or fewer, which makes the report work from only those rows and say so. In a distribution job, the report's live report file fails its run, with the reason in the step log. Cube crosstabs are not limited this way in this version. |
| Rows on the sheet | 1,000 rows, columns A to AZ, in the web report studio and viewer | A list taller than the sheet shows the rows that fit, keeps its Total under them, and says so ("shows 993 of the first 1,000 rows of its source, and its source has more rows"; its Total covers only the rows shown). The viewer and the downloaded file say it. A distribution job writes to Excel's whole sheet instead: every row goes out with its Total, and a report that would pass Excel's last row fails its sheet ("Nothing was distributed") rather than being cut. |
| Size of a saved report | 25 MB (as stored, 1 MB = 1,000,000 bytes) | Save refuses, and tells you the report's size and the limit. Nothing is saved and, for a report you are editing, your changes stay on screen: take a sheet, a picture or a large list out of the report, then save again. |
A distribution job is a template that makes outputs#
Do not think of a distribution job as a file. Think of it as a template that creates outputs.
A finished PDF or a sent email is a result. In Reportworq, the distribution job is the thing that produces those results, over and over. It is a reusable asset with levers built in: parameters, variables, queries, formatting, messaging, and distribution rules. Pull those levers, and one job can generate hundreds or thousands of personalized outputs.
Different parts of the same source workbook do different work. Some parts exist to build the output file. Some exist only to build a rich email message. Others make distribution decisions, and others exist to collect data back into the planning system. Together with the job's settings, the workbook becomes a specification for how information should flow through the organization.
This is where new users most often get stuck: they keep thinking of a report as a document to send rather than part of a reusable system. Once you build one solid master workbook that a job can slice, personalize, and repeat, you stop maintaining a separate file per audience, and that is the moment the product pays off.
The terms that follow#
Everything else is a name for one of the levers on that template.
Distribution job (job). The unit an author builds in the Distribution Job Editor: a single runnable, schedulable object that binds its reports, parameters, optional bursting, output formats, and delivery targets together. The author builds it once; Reportworq runs it every time it is scheduled or started.
Report and worksheet. A report is the content a job distributes, added on the job's Reports step: usually an Excel workbook, and also a web report, a PowerPoint deck, or a PDF. Where it helps to be clear, these guides call it a source report. For a workbook you also choose the specific sheets the job outputs, in order. The source workbook is where your existing Excel work lives, Reportworq automates it rather than replacing it. See Choose the source workbook and worksheets.
Output. What a run of a job produces and delivers: the PDF, the workbook, the PowerPoint deck, the email.
Parameter. A named input that drives the job. Change a parameter such as period, store, or cost center, and the formulas in the source workbook recalculate to that value. Parameterizing a job is what lets one template serve every audience. See Parameterize a distribution job.
Bursting, replication, and burst sets. Bursting is the outcome: many personalized outputs from one job, one per value, each going to its own recipient. One cost-center job becomes one packet per cost-center owner.
Two words describe how you get there, and they are not interchangeable. Replication is a parameter with several values fanning the output out by itself; set the parameter's replication mode and you are done. A burst set is the separate Bursting step, unlocked by flagging a parameter Burstable, which gives you per-file control over the values and the recipients.
You do not need Burstable to produce many outputs. A multi-value parameter is already replicating. Flag Burstable only when you want the Bursting step itself. See What bursting is and when you need it.
Distribution. The output formats a job produces (PDF, Excel, PowerPoint, CSV, and more) and the destinations it delivers them to (email, network folder, SharePoint, Slack, Teams, and others). One job can produce several formats to several places at once. See Distribution overview.
Workspace. The governed content space your jobs, reports, and campaigns live in, with its own folder tree, items, contacts, and security. The Workspace is also the home screen you browse. A server can host more than one workspace.
Most organizations need only one. Folder-level security handles separating teams inside a single workspace, which is simpler to run than several. The clearest reason to add a second is credentials: a workspace holds its own data-source connections, so when two groups must reach the same system under different credentials, giving each its own workspace is the way to do it. Splitting purely to separate content is usually better served by folders.
Contribution. The reverse direction. A contribution campaign turns a workbook into an input form, hands it to the people who own the numbers, collects what they enter, routes it for approval where you need one, and writes the approved values back to the planning system. The pieces are the campaign (the round of collection), the form each contributor fills in, and the approval step that gates the write-back. Distribution sends information out; contribution brings it back in, through the same security and the same audit trail. See Contribution overview.
AI context. Reportworq's AI answers from the outputs of the jobs you have already built and approved, rather than querying your systems directly. That keeps answers traceable to something a person signed off, and keeps a user's AI access identical to their normal access. An author steers it per job with an AI profile, which tells the AI what the job's output means and how to talk about it. This is the steering wheel that stops answers drifting. See Create an AI Profile for a distribution job and AI in Reportworq.
The value loop#
These pieces serve one repeating cycle: an author defines a distribution job once, Reportworq runs it (on demand or on a schedule) by refreshing the data and rendering each output, and then distributes the results to where people work. Because the job is defined once and reused, the authoring effort carries over to every future run. The full arc, from authoring through monitoring, is walked through in The distribution job lifecycle.
Where to go next#
- Build reports and data models: the Reporting and Data Modeling guide.
- See the arc in action in The distribution job lifecycle.
- Build your first job in Your first distribution job, end to end.
- Look up any term in the Glossary.
Feedback on this page
Comments, questions, requests, or something missing or unclear? Email us - the page you are on is filled in for you.
Email feedback on this pageOr write to support@reportworq.com directly.