Your ERPNext Projects Deserve Better Than Spreadsheets

Published On: 26 October 2026.By .
Project Management/ERPNext/9 min
ASSEMBLED BY HAND Spreadsheet Chat thread Timesheet export Standup notes ? 40 minutes later already out of date LOOKED UP One project record work, hours, risks, reports Status already on screen no assembly
FIG 00  The same question, answered two ways. The difference is not effort, it is where the data lives.
The short answer

Project delivery usually ends up outside the ERP: work items in one tool, hours in a timesheet export, risks in a conversation nobody wrote down. The cost is not the second licence fee, it is that nobody can answer "where are we?" without assembling it by hand. Auri PMS puts tickets, sprints, releases, time logs, risks, dashboards and client status reports on the same ERPNext Project record the rest of the business already uses. It runs on ERPNext v16, is free and open source on the Frappe Cloud Marketplace, and modifies nothing in ERPNext core.

What matters most
  1. The problem is not a missing dashboard. It is that delivery data and business data live on different records, so every answer needs reconciling first.
  2. A health traffic light somebody sets by hand is an opinion. Scored from the project's own data, it is a measurement, and it can say why.
  3. Weekly client reporting is where scattered data costs the most, because it cannot be skipped and cannot be automated when the inputs are in four places.
  4. Anything added to a production ERP has to leave core untouched, or you pay for it at the next upgrade.

It is four o'clock on a Friday. A client sends one line: "Where are we on the release?"

Somewhere a project manager opens a spreadsheet, then a chat thread, then a timesheet export, then a half-remembered standup note, and starts stitching an answer together by hand. Forty minutes later a reply goes out that is roughly accurate and already slightly out of date.

That scramble is not a people problem. It is a tooling problem. The work is tracked in one place, the hours in another, the risks in someone's head, and the client report in a deck rebuilt every week from four sources that do not agree with each other. The fix is not another tool to check. It is one fewer.

01

The question behind the question

"Where are we?" is really four questions wearing one coat. How much is done. How much is left. What is going to go wrong. And what of that is our fault rather than theirs.

Each of those lives in a different place in most setups, which is why the answer has to be assembled rather than read. The work breakdown is in the project tool. The effort is in timesheets. The risk is in a conversation. The client dependency is in an email thread from Tuesday that nobody tagged.

Status should be something you look up, not something you assemble.

02

Why delivery data drifts out of the ERP

Usually because the ERP was chosen for finance, inventory and operations, and project delivery was never part of that conversation. A dedicated project tool gets added later, often by the delivery team rather than by IT, and it works well enough on its own terms.

The problem shows up at the joins. Hours logged in the project tool have to be reconciled against the ones that reach invoicing. The project a client is billed for and the project a team works in are two different records with two different names. Reporting means exporting both and lining them up in a spreadsheet.

Two records

Delivery runs in one system, the business runs in another. Every report begins with reconciliation, and every number has two versions until someone decides which one is true.

One record

Work items are logged against standard ERPNext Projects. The project finance sees and the project the team works in are the same record, so there is nothing to reconcile.

03

What visibility actually means

Visibility is not a dashboard with more charts on it. It is whether the things that quietly go wrong announce themselves without anyone having to ask.

Most delivery problems are not hidden by bad people. A blocked ticket sits quietly for a week. A risk everyone knows about never gets an owner. A client approval holds up three tasks and nobody can say for how long. None of that is concealed. It is just not counted anywhere.

Four mechanismsWhat counting it looks like in practice

What goes wrong quietlyHow it gets counted
Work waiting on the clientA ticket marked Dependent on Client records a note and a waiting-since date. The wait is counted on every dashboard and listed in the client report, so it stops being read as your delay.
Risks with no ownerProbability and impact are each scored 1 to 5 and the product sets the level. Every risk carries an owner, a mitigation plan, a contingency plan and a due date.
Blocked work moving anywayStatus transitions are configurable per project role. A move the workflow does not allow is refused on the board rather than quietly accepted.
Effort overrun found lateHours are logged against a ticket with an activity type and a billable flag, then read against the estimate. Limits per entry and per day, plus lock days, keep the numbers trustworthy.

ScoringRisk levels, deliberately boring

Probability times impact, with the level following from the number rather than from whoever is loudest in the room. Open risks pull the project health down automatically, and risks flagged for the client report appear there with their mitigation.

Impact ↓ / Prob →12345
5510152025
448121620
33691215
2246810
112345

Low is 1 to 4, Medium 5 to 9, High 10 to 14, Critical 15 to 25. The thresholds are configurable, which matters more than the defaults do.

04

Five dashboards, five questions

A dashboard is only useful if it answers a question someone is actually asking. Five come up on nearly every delivery team, and each belongs to a different person.

DashboardThe question it answersWho it is for
Project HealthIs this project on track, at risk or off track, and why?Project managers
PortfolioHow do all my projects compare on budget burn, overdue work and capacity?Management
My WorkWhat should I focus on today?Every team member
QualityAre we opening bugs faster than we fix them?QA and engineering leads
Client Status ReportWhat do I tell the client this week?Account and delivery teams

They share one theme and one range, so switching from one month to overall changes all of them together rather than one at a time. Every count on every dashboard is clickable and opens the tickets or time entries behind it, which is the difference between a number you believe and a number you can check.

Auri PMS Portfolio dashboard comparing every project on open tickets, overdue percentage, completion, budget burn and billable share
FIG 01  Portfolio. One row per project, every count clickable through to the tickets behind it.
Auri PMS My Work dashboard showing a personal focus list, tickets grouped by due date, hours logged against weekly capacity and active sprints
FIG 02  My Work. The same data as the portfolio, cut to one person and one day.
SIX LIVE CHECKS Overdue share of open tickets Sprint burndown against the ideal line Hours spent against work completed Release progress near its due date Open risks and overdue mitigations Created against resolved, last four weeks SCORE starts at 100 minus 10 amber, minus 25 red On Track nothing flagged At Risk one amber or more Off Track one red or more
FIG 03  No manual input anywhere in the chain. Every threshold is configurable.
05

Scoring health instead of declaring it

The health verdict is the part worth being careful about, because a traffic light somebody sets by hand is an opinion with a colour attached. Here it is calculated live from the project's own data, and each check that fires costs points and is written out as a reason in plain words, such as "12% of open tickets are overdue (3)."

The thresholds below are defaults rather than laws. They are the first thing to tune once a team has run on them for a month, because a studio doing two-week sprints and an implementation team doing six-month rollouts do not agree on what late means.

CheckAt risk (amber)Off track (red)
Overdue open ticketsMore than 10% of open ticketsMore than 25%
Active sprintSlightly behind the ideal burndownWell behind it
EffortHours run ahead of work completedHours exceed the estimate
ReleasesAt 80% or less and due within 7 daysPast the due date and under 80% done
Risk registerA High open risk, or one past its mitigation dateA Critical open risk
BacklogMore created than resolved over four weeksNot applicable
The verdict is optional, the facts are not

Showing the health status on the hub, on the portfolio and in the client report are three separate switches, and all three are off by default. Counts and charts always appear. That lets a team keep the judgement internal while the client sees only what happened, which is usually the right order to introduce it in.

06

Client reporting without rebuilding the deck

Weekly client reporting is where the cost of scattered data lands hardest, because it is the one task that cannot be skipped and cannot be automated when the inputs live in four places.

When the delivery data already sits in one system, the report stops being an assembly job. Delivered work, work in progress, what is waiting on the client, hours, the plan and the open risks are all read from the same project record.

  • Pick a layout, or use the built-in one. Custom email templates work once enabled
  • Switch sections on or off and drag them into order, for one send or as the default
  • Remove a ticket from this send, or add one with a note for the client
  • Preview the report exactly as the client will receive it
  • Email it, download the PDF, or schedule it weekly, monthly, or both
  • Every send is logged with delivery status, so there is a record of what went out and when
Auri PMS Client Status Report preview showing delivered work, work in progress, items waiting on the client, hours and open risks
FIG 04  The client status report, previewed exactly as it will be emailed or exported to PDF.

The goal is not more reporting. It is a project where the report writes itself.

07

How a project moves through the system

From first ticket to final report, the same data carries the project. Nothing is retyped between stages, which is the entire point.

01 Plan tickets, sprints, releases 02 Work board moves, time logged 03 Watch health, risks, client waits 04 Compare portfolio, capacity, quality 05 Report client status, on a schedule one project record underneath all five
FIG 05  The report at the end is built from the same records the work started in.

Daily due-date digests keep the team moving between reports, and optional email-to-ticket intake turns an incoming message into a work item without anyone rekeying it.

08

Getting started without a rollout plan

Adding anything to a production ERP usually means a project of its own. This one does not. The only setup the app requires is adding project members, and ordinary ERPNext projects are left alone until you tick the box that brings one in.

  1. Enable it on one project

    Open an existing ERPNext Project, go to the Auri PMS tab, set a short identifier such as AURI and tick Enable. Ticking it selects every active ticket type; untick the ones you do not want. Pick a real project that is already running, not a test one, because the dashboards only say anything useful with real work in them.

    PASS the project appears on the Project Hub
  2. Add members and set roles

    This is the only setup the app insists on, and it controls who sees the project. Three roles exist at the system level: Admin, Manager and User. Inside a project, each member's project role carries nine capability flags covering who can create, edit, delete, manage releases and members, log time, edit other people's time and be assigned.

    PASS each member sees the project, and nobody outside it does
  3. Define your modules before logging tickets

    Modules are the parts of the product a ticket belongs to, such as Sales, Stock or Accounts. Do this first. Dashboards, reports and the client report all break their numbers down by module, so tickets logged before the modules exist produce a breakdown with a large unhelpful "none" bucket in it.

    PASS the Module field offers your own list
  4. Run one sprint and log time against it

    Add tickets, set a goal and dates, press Start. Agree the time rules up front: a limit per entry, a limit per day, and a lock day after which old entries cannot be edited. Set each member's allocation as a share of weekly capacity so over-allocation and burnout flags have something to measure against.

    PASS the burndown moves and My Work shows hours against capacity
  5. Log your real risks, then leave the thresholds alone for a month

    Score probability and impact, assign an owner, write the mitigation and a due date. Flag anything waiting on a client answer as Dependent on Client with a note. Resist retuning the health thresholds in week one. Give it a month of real data first, then adjust in PMS Settings under Risk and Health.

    PASS the health banner shows a status with reasons you recognise
  6. Send one client report by hand before scheduling it

    Choose a layout, set the sections and their order, preview it, and send it manually once. Read it as the client will. Only then set the weekly day or monthly date, and add the recipients on the project's Auri PMS tab. Scheduled sends use the defaults in PMS Settings, so get those right before automating.

    PASS the send is logged as a Project Status Report with a delivery status
Auri PMS team capacity view showing each member allocation percentage, logged against allocated hours, and over-allocation and burnout flags
FIG 06  Allocation set in step four is what gives the over-allocation and burnout flags something to measure against.
09

Does it change ERPNext itself?

This is the question that decides whether anything gets installed on a production ERP, and it deserves a direct answer: no.

ConcernHow it is handled
Upgrade safetyNothing in Frappe or ERPNext core is modified. Everything is added through hooks, custom fields, custom permissions and document events
Learning curveStandard Desk UI only. Forms and lists keep the familiar Frappe look, so there is no second interface to train people on
HostingNo OS packages and no custom bench commands at runtime, so it installs on a managed Frappe Cloud site as readily as a self-hosted bench
ReversibilityUninstalling removes the custom fields, the Kanban board, the workspace and the permissions it added. ERPNext activity types are left alone
LicenceMIT, with the source public. Free on the Frappe Cloud Marketplace, with no paid tier behind it
VersionFrappe v16 and ERPNext v16, with ERPNext as a required app
10

Where this falls short

Worth being clear about the edges before you put it on a live client project.

  • It needs v16. Frappe and ERPNext v16 are both required. If you are on v15 and not planning to move yet, this is not the quarter for it.
  • The dashboards are only as honest as the logging. Health is scored from tickets, hours and risks. A team that logs time on Friday for the whole week, or never raises a risk until it has already happened, will get a reassuring green light that means nothing.
  • The default thresholds will not suit everyone. Ten percent overdue is a sensible starting point, not a universal truth. Expect to tune them, and expect the first month of scores to be noisier than the second.
  • A health score is a summary, not a judgement. It tells you where to look. It does not tell you whether the client is happy, whether the scope was right, or whether the team is about to quit.
The point was never the dashboards. It was the shape: planning, work, hours, risks and client updates describing the same project instead of five versions of it. Projects rarely fail in one dramatic moment, they drift, and the teams that stay ahead are not the ones holding more meetings. They are the ones who can see the drift while it is still small. Start with one project that is already running, give it a month of honest logging, and see whether the Friday question still takes forty minutes.
11

Frequently asked questions

Does this replace ERPNext Projects?
No. It builds on the standard Project record rather than replacing it. Work items are logged against existing ERPNext Projects, so the project a finance team sees and the project a delivery team works in are the same record.
Who can see a project once delivery data sits in the ERP?
Access follows project membership. Team members see the projects they belong to plus any marked public, managers see the projects they manage, and administrators see everything. Private projects stay hidden from people outside them, so client-sensitive work is not exposed by moving it into a shared system.
Can billable and non-billable time be tracked separately?
Yes. Each time entry carries an activity type and a billable flag, and limits per entry and per day keep the data trustworthy. Billable share then appears alongside progress and budget burn, so commercial health and delivery health are read from the same screen.
How is project health calculated rather than guessed?
Health is scored live from the project's own data using six checks: overdue open tickets, sprint burndown against the ideal line, effort spent against work completed, release progress near due dates, open risks, and backlog trend. The score starts at 100 and loses points for each check that fires, and every flagged check is listed in plain words.
Can clients receive status reports automatically?
Yes. A report can be previewed, exported to PDF or scheduled weekly or monthly, with sections switched on or off and reordered. Every send is logged with its delivery status, so there is a record of what went out and when.
Does installing it modify ERPNext itself?
No. Everything is added through hooks, custom fields, custom permissions and document events rather than edits to core. That keeps future ERPNext upgrades straightforward, which is the main reason teams hesitate to add anything to a production ERP.
Which ERPNext version does Auri PMS need?
Frappe v16 and ERPNext v16, with ERPNext as a required app. It uses the standard Desk interface only and needs no OS packages or custom bench commands at runtime, so it installs on a managed Frappe Cloud site as readily as on a self-hosted bench.
Where this goes next

Want to see it on your own project data?

Auri PMS is free and open source on the Frappe Cloud Marketplace, so you can install it and judge it yourself. Auriga IT implements and supports ERPNext across manufacturing, distribution and services. If you are weighing whether delivery belongs inside your ERP, we are happy to look at one project with you.

Install from the marketplace → Talk to our ERP team →
Sources

Verified 26 October 2026 against the Frappe Cloud Marketplace listing and the Auri PMS repository README. Feature names and thresholds reflect the current release and may change in later versions.

  1. Auri PMS on the Frappe Cloud Marketplace
  2. Auri PMS source and documentation on GitHub

Related content

Stay Close to What We’re Building

Get insights on product engineering, AI, and real-world technology decisions shaping modern businesses.

Anoop Verma
Anoop Verma
Anoop Verma is a Functional Consultant and Business Analyst at Auriga IT with 5+ years of experience implementing ERPNext across manufacturing, textile, and retail businesses. He is a Frappe-certified ERPNext Functional Consultant and SAP MM certified, specialising in Buying, Selling, Stock, Manufacturing, Subcontracting, and Accounts.
Go to Top