Ask the plant head at almost any mid-size manufacturer whether their ERP works, and you’ll hear some version of “yes, mostly”. Purchase orders, inventory, invoicing, GST and the general ledger all run through SAP, Oracle, Microsoft Dynamics or Tally, and nobody wants to go back.
Then walk the floor. Gate entries are written in a register. QC checklists are on paper and get typed into Excel at the end of the shift. The sales team keeps distributor claims in a shared spreadsheet, and someone in planning spends every Monday morning stitching three ERP reports together because the one they need doesn’t exist.
That is the other 20%. It rarely shows up in an ERP project plan, but it is where a lot of the daily friction, and a fair amount of the risk, actually sits.
Where the gaps usually are
The specifics differ by industry, but in manufacturing, FMCG and consumer goods the same handful of gaps come up again and again:
- Gate and weighbridge entry. Vehicle in, weight captured, material matched to a PO. This is often manual, and the data reaches the ERP hours later.
- Quality checks. Incoming material inspection, in-process checks and final QC, with photos, sign-offs and deviation handling. ERPs record the result; they rarely handle the workflow well.
- Batch and lot traceability. Answering “which customer got product from this raw material lot?” in minutes rather than days, which matters most during a recall or an audit.
- Dealer and distributor portals. Order placement, scheme and claim tracking, and secondary sales visibility. Few distributors will log in to your ERP, and you probably don’t want them to.
- Shop-floor data. Production counts, downtime reasons and shift handovers, captured on a tablet at the line instead of a clipboard.
- Management reports. The dashboard the MD actually asks for, which needs data from the ERP, the HR system and two spreadsheets.
None of these is a reason to replace the ERP. Each is a specific workflow with a specific owner, and each costs time every single day.
Why “just customise the ERP” often goes wrong
The obvious answer is to build the missing pieces inside the ERP. Sometimes that’s right. Often it isn’t, for three reasons:
- Upgrades get harder. Heavy customisation inside the core makes every version upgrade a project of its own. Many companies end up stuck on an old version because moving would break years of custom code.
- It’s expensive for what it is. ERP consultants charge ERP rates, and a simple mobile-friendly QC form can turn into weeks of work inside a platform that wasn’t designed for it.
- The users aren’t ERP users. Shop-floor supervisors, security guards at the gate and distributors need something simple that works on a phone, often in a regional language. Giving them ERP licences and training rarely ends well.
The opposite extreme is re-implementation or a switch to a “better” ERP. That’s a 12–24 month programme with real business risk, and the new system will have its own 20% gap on day one.
The third option: build alongside the ERP
The approach that tends to work is to leave the ERP as the system of record and build small, focused applications next to it that talk to it through integrations.
In practice:
- The ERP stays the source of truth for masters (items, vendors, customers) and for financial transactions.
- The new application owns the workflow. The gate app, the QC app and the distributor portal handle the screens, approvals, photos and notifications.
- Integration moves the data both ways. Masters flow out of the ERP, and completed transactions flow back in. This happens through the ERP’s APIs where they exist (SAP, Oracle and Dynamics all have them, and Tally can be integrated through its XML/ODBC interfaces) or through a small middleware layer where they don’t.
Because each application is small, it can go live in weeks rather than quarters, and the ERP upgrade path stays clean.
Three examples
Illustrative scenarios based on common requests. Real client examples to follow.
A packaging plant’s QC workflow. Inspectors used paper checklists that were typed into Excel the next day, so a failed batch could already be on a truck before anyone saw the result. A tablet app with checklists by product, mandatory photos and an automatic hold on failure solves this. Approved results post back to the ERP’s quality module, and a failure blocks dispatch for that batch.
An FMCG brand’s distributor claims. Scheme claims came in by email and Excel and were reconciled by hand each month, with disputes that dragged on. A distributor portal lets distributors submit claims with proof, checks them against the scheme rules automatically, and posts approved credit notes to the ERP. Finance sees every claim’s status in one place.
A components maker’s traceability. A customer audit asked for the raw-material lot behind a specific shipment. Pulling it together from ERP reports and production registers took two days. A thin traceability layer that links GRN lots, production batches and dispatches makes it a single search.
How to scope your own 20%
You don’t need a big programme to start. A useful first pass takes an afternoon:
-
1List the spreadsheets
Every register, Excel file and WhatsApp group a department relies on.
→ A list of gaps -
2Put a cost on each
Hours per week, plus audit, recall and leakage risk.
→ Cost per gap -
3Check the data path
What each gap reads from the ERP and what it writes back.
→ Integration effort -
4Pick one
Highest cost, simplest data path. Build that first.
→ Your first appBuild this first
- List the spreadsheets. Ask each department head which spreadsheets, registers or WhatsApp groups they rely on every week. Each one is a gap.
- Put a cost on each. Estimate the hours per week, plus the risk: audit findings, recall exposure, revenue leakage from unreconciled claims.
- Check the data path. For each gap, which ERP data does it need, and what does it need to write back? This tells you how much integration is involved.
- Pick one. Choose the gap with the highest cost and the simplest data path, and build that first. A working app on the floor will teach you more than a requirements document.
When re-implementation is the right call
To be fair to the other side: if the ERP can’t handle your core transactions, if it’s out of support, or if the 20% is really 50%, building around it is only postponing the decision. We’ll cover how to make that call in a later post on build vs buy vs extend.
For most mid-size manufacturers, though, the ERP is fine. What’s missing is the last stretch between the ERP and the people doing the work.
Book a free 30-minute ERP gap review. Bring your list of spreadsheets. We’ll go through which gaps are worth closing first and what the integration with your ERP would involve. Talk to us →