ERP Solutions

Business systems that fit how you operate.

Fit the week you already run, not a module brochure

Most firms that ask LavisTech about ERP are not trying to become a different company. They already take orders, move stock, raise invoices, and file GST. The pain is that those facts live in four places that almost agree: Tally for the books, Excel for the “real” stock, a WhatsApp group for dispatch, and a folder of photos when someone needs proof. Across India we see that shape in SMEs every month. An ERP is useful when one system of record can carry the operational week — and honest when it does not pretend the accounts package, the night shift, or the branch book will vanish because a salesperson showed a dashboard.

Fit means the screen a clerk uses on Tuesday matches how an order actually starts. It means item codes are the ones the store already shouts across the floor, not a new religion invented in a workshop. It means opening stock is a number someone will defend, not a cell that was typed once in March and never counted again. If those sentences make you uncomfortable, you are in the right conversation. If they sound like someone else’s problem, you may still want a product — and you should buy one, not commission a custom system to feel serious.

What “how you operate” looks like when we sit with it

We sit with the people who enter orders, not only with the person who bought the project. The order-entry person knows the exceptions: the customer who is billed as two parties, the job that substitutes a grade, the branch that sells from a van. The buyer knows the brochure. Custom ERP is the exceptions, encoded once, so the afternoon is not spent reconciling. Photograph the workbooks. Note who updates which tab, and which tab people actually trust. You will find opening stock in a branch book, item codes that collide, and a file named something like final_v3 that is the real price list. That pile is the data model. A workshop that ignores it produces a pretty empty system.

A package “BOM” assumes a finished good you repeat. A jobbing shop often has a recipe that changes per order — hold a batch, substitute a grade, bill the exception the customer already agreed. Multi-branch trade often means each branch almost has its own company. A service firm may bill time, parts, and retainers in a mix no product author put on a roadmap. Those are the firms for whom “fit” is not a slogan. If you are standard wholesale with standard GST and a willingness to change a form, fit might mean changing yourselves to match a product. That is a respectable choice. It is a different job than the one on this page.

Tally can stay for the books — and often should

Accounts know Tally. Auditors know Tally. Ripping it out on day one so the new screens can “do GST” is how you get a go-live that finance refuses. A nightly or on-post integration — order and invoice in the custom system, books in Tally — is a phase, not a failure. You can move the books later if you still want to. Most SME manufacturers we talk to never need to. The custom part is then the jobbing rules, the branch stock, the dealer portal, the thing Tally will never grow. You are choosing where the system of record for operations lives. You are not buying a religion about ledgers.

GST history still matters. Five years of invoices you must keep do not become someone else’s problem because you bought new screens. Filing season is when you earn finance’s look, or lose it for a year. If Tally remains the place a chartered accountant will open, design the integration so a posting or a file they already understand arrives without a hero in accounts retyping. If the hero is still retyping after go-live, you have a reporting skin. We will say so. Keeping Tally is not a lack of ambition. It is refusing to fail the one department that can stop the project with a single “the numbers are wrong.”

Custom versus SAP versus Odoo, without a logo contest

SAP is for organisations that will change themselves to match a global product and can pay for the implementation machine that comes with it. Most fifty-person plants in Kerala are not that buyer, however often the logo appears in a pitch. Odoo can be a good fit if you stay close to standard apps — sales, inventory, invoicing — and you have a partner who will say no to “just a small custom module” every week. Odoo goes wrong when those small modules become a private fork you cannot upgrade. Custom is for when your billing, jobbing, or multi-branch rules are why customers stay, or when the last package quote to “make it work like we work” already looks like a unique system with someone else’s name on the invoice.

The mistake in the first meeting is comparing a licence slide to a custom quote. Implementation, data, training, and the “small customisations” are the real number on every side. We would rather you buy Odoo and stay standard than commission custom because the word sounds serious. We would rather you commission custom than spend two years customising a package into a corner. The comparison of those three paths, with the failure modes attached, is written as custom ERP versus SAP versus Odoo. Read it before you take three meetings that are not about the same thing. If a vendor quotes a full ERP from a phone call, they are quoting a product shape, not your process.

Implementation of a package is not the same job as a bespoke build

People use “ERP implementation” for both, and then wonder why two quotes cannot be compared. Implementing a package means configuring modules, writing the gap between “the app exists” and “this is how we invoice retainers,” cleaning data, training, and living near the product’s process. A bespoke build means designing the data model around your exceptions, writing software you own, and still doing the same cleaning, training, and parallel run. The months of data and habit appear on both sides. The difference is where the money goes after year one: licences and partner days, or hosting and change you approve. Calling a heavily customised package “implementation” does not make you own it.

Package calendars are not automatically shorter. Odoo or SAP still need item codes that three people agree, staff who will click the live system, and a stretch where two systems are true. Custom “three months for a slice” assumes those people exist too. Lie about any of that and the months reappear, with blame attached to the software. If you are still deciding whether the job is a product at all, that is a consulting question before it is a build. We run that shape as IT systems consulting when two packages already disagree, or when nobody in the room can name the system of record. Do not skip it to look decisive.

Reporting people will defend, not a dashboard they screenshot

A report is useful when it is the same facts as the invoice, the stock move, and the job — entered once. It is decoration when people export to Excel to “do the real work.” Directors ask for twelve charts in January. What they need on a Thursday is yesterday’s order, this morning’s stock at one branch, and a GST view that does not fight Tally. We build those first. Fancy tiles can wait until the numbers have survived a parallel run. If after go-live the WhatsApp group is still the dispatch list, you do not have a reporting problem. You have a path problem. Fix the path, or admit the project was a screen.

Accounts will ask whether they can see what they already see in Tally, only sooner. Operations will ask whether the stock number matches the rack. Those are two different questions. A single “MIS dashboard” that pretends to answer both usually answers neither. We would rather ship two boring lists that match the documents than one colourful page that cannot be audited. When a director wants last year’s trend on day one, we write it on the out-list. History you must keep is a data migration job with its own checks. It is not a chart colour.

The first slice, written so narrowly it feels rude

Not “go live on the first of the month with everything.” A first live slice is usually orders and stock for one location, or jobs and time for one team. Accounts often stay in Tally for that phase. Production, planning, and the customer portal come after the numbers are trusted. A paragraph you can paste into a statement of work looks like this: one branch; orders and stock; yesterday’s order is findable; Tally receives a posting or a file we agreed; parallel run until the numbers match for ten working days; no production planning, no customer portal, no second branch; change requests written down before they are built. “Full ERP this year” is not a calendar. That paragraph is.

Live means yesterday’s order can be found by the person who will be asked about it, without the old spreadsheet. It does not mean every report mentioned in a kickoff. When the numbers match for a stretch you agreed in discovery — often two to four weeks of parallel run — you switch the old path off. Switching off is a management act. Software cannot retire a workbook people still trust. Stop a phase when the slice does what the sentence said and the next request is a new sentence. Combining them “while we are here” is how first slices become nine-month programmes with no go-live. The longer definition of the word, and when a product is the better answer, is in what custom ERP is.

Where the months actually go

Discovery is two to six weeks if we can sit with the people who type and see the embarrassed workbooks. A first slice is often eight to sixteen weeks of build, migrate, and train — three to six months on a wall calendar once you add waits you do not control. Parallel run is two to six weeks and is the phase sales Gantt charts skip. A fuller system with stock, jobs, accounts, and reporting is nine months to well over a year, usually in phases. Anyone promising “full ERP, one go-live, ninety days” is pricing a different building than yours. The phase table and the hidden weeks — Tally at month-end, item codes that three people must agree, a weighbridge vendor’s test key — sit in ERP implementation cost and timeline.

We will not print a rupee total that will be wrong for your connector, your branches, or your GST exceptions. What we will say: a focused slice with few integrations and a clean-enough item list is a medium custom-software job; a programme that replaces several spreadsheets and talks to Tally, a site, and a second location is larger and should be phased. If two vendors differ by a factor of three on the same one-page brief, one of them has not priced migration, parallel run, or handover. Ask them to show which. Staff time is not optional. An operations lead, a day a week. Accounts, enough hours to say the numbers are wrong. The branch that will click it live. Without those names, discovery should not end.

Opening stock, item codes, and the Monday you do not want

Opening stock is where implementations go to die, quietly. Someone types a number from a book. Someone else knows the book was last counted before Onam. The new system goes live with confidence and a lie. We would rather delay a week for a count you will defend than ship a dashboard that will be argued about until the next financial year. Item codes that are not unique are the cousin of that lie. Three spellings of the same customer, two codes for one SKU, a “MISC” that hides a fifth of the value — that is not a data-entry footnote. It is the project. Budget the cleanup as work, or budget the surprise.

WhatsApp is often the real dispatch list. If staff will not leave the group, the system is optional, whatever the steering slide said. The test is not “they attended training.” The test is whether a Thursday dispatch is created in the system without you in the room. If it is not, do not add a customer portal. Ask whether undo was missing, whether the phone was wrong, or whether the incentive still lives in the group. We have no interest in a go-live photograph. We have interest in the week after, when the hero in accounts is on leave and the numbers still match.

When we will tell you not to commission custom

If the pain is still one person’s afternoon, software will not pay back. If sales, stock, and invoicing are ordinary and a partner will refuse weekly custom modules, buy a product and stay near standard apps. If a group company is mandating SAP and funding the programme, that is a different conversation than an Indian SME inventing “our own ERP” because the phrase sounds like growth. Building Tally again with prettier screens is how custom projects get a bad name. We will not take that brief and call it transformation.

If you cannot name a first-slice sentence, you are not ready to build. You are ready for a short discovery or a systems review. If a director wants every module because a competitor has a slide, write the expensive assumption instead: staff will record this job here instead of the sheet. Build only what that sentence needs. Reports can wait. The second warehouse can wait. That discipline is how you still have a company to implement the next slice. Bring the unofficial workbooks and the login complaint, not a mood board. If that is the conversation you want, start on the contact page with the sentence and the systems that must stay.