Rebuilding Shift 2 From Paper Every Morning? 7 Signs You Need Custom Software in Manufacturing

MSMEs, the micro, small and medium enterprises behind India's factory floors, produce 36.2% of the country's manufacturing output (2021-22). Much of that output still runs on Excel and paper. Audits of real spreadsheets found errors in 94% of the sheets reviewed.

• Can you tell your MD what shift 2 produced before shift 1 walks in tomorrow?
• If an OEM flags one bad batch today, can you prove its inspection history before lunch?
• How many times does your team type one job card into Excel, Tally and the dispatch sheet?

If you nodded at any of these, the problem is rarely your people. It is a setup that has outgrown its tools. Edhaas Digisoft, a Pune-based custom software firm for auto-component and manufacturing MSMEs, wrote this guide from the shop-floor side: seven signs, the numbers behind them and how we fix each one.

What are the signs you need custom software in manufacturing?

The signs you need custom software in manufacturing are repeated workarounds your team now treats as normal: Excel as the main system, shift data on WhatsApp, the same entry typed in several places, slow batch proof, and standard software that forces you to change your process. Seven signs follow.

When we start with a new plant, we begin at the shift handover. Most of these signs show up there first, long before anyone calls them an IT problem.

The seven signs, in the order we usually see them:

The 3-minute Shift 2 Test

Score one point for each sign you recognise. Zero to two: tighten your current sheets. Three to four: pick one process and digitise it first. Five or more: a custom build deserves a serious look.

The first two signs are the quietest, so we start there.

Why do Excel sheets and WhatsApp reports stop working on a shop floor?

Excel sheets and WhatsApp reports stop working on a shop floor once the number of formulas, shifts and people grows past what one person can check. Errors become near certain, data arrives late, and nobody can trace who changed what. Here is the maths behind it.

Raymond Panko, a professor at the University of Hawai'i, studied this for years. His error table shows that if each formula is right 98% of the time, a sheet with 100 formulas has an 86.7% chance of at least one error.

"...even reducing errors by 80% to 90% is extremely difficult..."

Raymond R. Panko, Professor, University of Hawai'i (source)

Takeaway: A sheet with 100 formulas is almost certain to hold an error, even when each formula is 95% to 98% accurate.

Alt text: Bar chart showing that the chance of at least one spreadsheet error rises from 18.3% to 86.7% at 98% formula accuracy as formulas grow from 10 to 100.

Source: Panko, "Thinking is Bad", University of Hawai'i. Values rounded.

WhatsApp adds a second gap. A photo of a handwritten sheet is a picture, not data, so nobody can total it, filter it or trace it later.

When we build this for a forging line, we map the shift handover first. We list what the supervisor writes at each shift change, then turn those fields into a digital production tracking form that operators fill on a phone or tablet and that feeds one live dashboard.

Fixing entry at the source still leaves another habit: typing the same data again somewhere else.

How does repeated data entry across production, quality and stores hurt a plant?

Repeated data entry hurts a plant because each retyped job card is a fresh chance for a wrong number, and each separate record leaves production, quality and stores holding different versions of the same truth. Management then sees stale or conflicting figures instead of a live plan versus actual view.

A job card is the sheet that travels with a lot and records what was made at each stage. The problems below show up wherever that sheet gets retyped.

Problem 1: One job card, three entries. Operators write it, the office types it into Excel, and accounts retype it into Tally.

Fix: One entry at the machine, linked to stock and dispatch. Tally or your ERP (Enterprise Resource Planning software for finance, stock and purchase) stays in place through an integration.

Problem 2: Three teams, three versions. Production, quality and stores each hold their own file, so numbers differ by department.

Fix: One shared batch record with role-based access, so each team edits only its own fields.

Problem 3: Plan versus actual arrives next morning.

Fix: A live plan vs actual dashboard that shows downtime, rejection and OEE (Overall Equipment Effectiveness, a score of how much planned time makes good parts at full speed).

We start with the entry typed most often, usually the job card. Then we link it to Tally or your ERP, so nobody has to replace systems that already work.

Disconnected records are annoying on a normal day. They become expensive when a customer calls about one batch.

Why does slow batch traceability put audits and OEM orders at risk?

Slow batch traceability puts audits and OEM orders at risk because customers and auditors expect proof of what was made, inspected and dispatched, and paper records make that proof slow to find. IATF 16949, the automotive quality management standard, adds customer-specific requirements that suppliers must be able to show.

OEM stands for original equipment manufacturer, the vehicle maker. A Tier-1 is a direct supplier to that OEM, and Tier-2 units supply the Tier-1. Buyers at each level ask for PPAP files (Production Part Approval Process) and APQP records (Advanced Product Quality Planning), both published by AIAG, the Automotive Industry Action Group.

A disputed batch is where slow records turn into real cost. Poor quality, including scrap, rework and complaints, takes a measurable share of sales.

Takeaway: The average manufacturer loses around 15% of sales to poor quality, and complex products can reach 35%.

Alt text: Bar chart showing the cost of poor quality in manufacturing at 5% of sales at the low end, about 15% on average and 35% at the high end.

Source: IISE, "Measuring the cost of quality".

"...ranges from 5 percent to 35 percent of your sales dollar."

Institute of Industrial and Systems Engineers (IISE), on the cost of poor quality in manufacturing (source)

When we build batch traceability for a machining shop, we link each inspection result to the batch number, invoice and dispatch record. One search then returns the full history, which is what an auditor or an OEM buyer actually asks for.

Once the pain is clear, the next question is which route fits your plant.

Have an audit date or a disputed batch on your desk? Schedule a Consultation and we will map one process with you first.

Custom software vs Excel vs off-the-shelf ERP: which suits a manufacturing MSME?

Custom software suits a manufacturing MSME when its process is specific, Excel suits only very small or short-term needs, and off-the-shelf ERP suits plants whose process already matches the package. The table compares fit, data entry, audit proof, change cost and subscription load.

Per-user subscriptions also grow as you add operators, supervisors and inspectors. Before you decide, compare one build quote against the subscription total over several years.

No tool removes human error on its own. Panko's research notes that packages do not automatically reduce errors, because people still make decisions. What a purpose-built system removes is retyping, and it makes mistakes visible sooner.

Before we quote, we ask for your process: handover sheets, inspection formats and dispatch papers. A price without that step is only a guess.

If the table points you toward a build, the next question is how to start without stopping production.

How do you move from Excel to custom software without stopping the line?

You move from Excel to custom software without stopping the line by digitising one process at a time, running the new screen beside the old sheet for a short period, and switching only after the numbers match. Our four steps are below.

The four steps we follow are:

Modular rollout keeps the budget close to what one process needs. More modules follow only when the first one has proved itself.

Here is what that looked like in two real builds.

What Most Guides to Custom Software Miss: The Shop-Floor Mechanism

Most guides on custom software for manufacturing miss the shop-floor mechanism: what actually changes at the machine, at the inspection table and at dispatch. They list generic signs, but they skip how data moves between shifts, who approves what, and how a batch is proven later.

Edhaas Digisoft, a Pune-based custom software firm for auto-component and manufacturing MSMEs, measures results in terms a Plant Head or MD uses: cycle time, retrieval speed and dispute handling. Two builds show it.

Edhaas Digisoft client data: Production log sheets

Challenge: A manufacturing company faced delays from manual production tracking.

What we built: A digital log sheet system with real-time shop-floor data entry and dashboards.

Results: 40% improvement in production cycle time. Zero delays in reporting and compliance. Simpler process tracking.

Edhaas Digisoft client data: Final inspection and dispatch clearance

Challenge: Final inspection relied on manual reports and physical signatures. A customer reported a defective item from a QA-checked batch, and manual record retrieval slowed the response.

What we built: Digital inspection reports with electronic approvals, inspection data auto-linked to invoices and dispatch documents, and a digital Certificate of Analysis (CoA) issued with every batch.

Results: The disputed batch was verified in minutes, confirming scrapped items and protecting client credibility. Zero missing records with audit-ready documentation. Improved customer confidence.

In both builds, the gain came from the rules behind the screen: who approves a record, and which invoice and dispatch record it links to.

Quick answers to common questions follow.

Conclusion

If your plant runs on Excel, WhatsApp and retyped job cards, the signs are already on your shop floor. Each one costs time, and the later ones cost audit confidence and customer trust.

You do not need to digitise everything at once. A custom production management build can start with the process that hurts most, such as shift logs or batch proof, and add the rest later.

Ready to map your first process? Schedule a Consultation with the Edhaas Digisoft team.