Which ERP Apps to Turn On First (and Which to Leave Off for Six Months)

Almost every first conversation about an ERP starts the same way. The founder has seen the app grid, counted forty or so tiles, and asks which ones they need. The honest answer is fewer than they think, and the order matters more than the number. Turning on too much at once is the single most common reason a first ERP build drags past its date and never quite gets adopted by the team.

So here is the order we use when a small or mid-sized business in the Gulf, Europe or North America asks which ERP apps to start with. It is built around one rule: an app only goes live when the process behind it is already stable on paper.

Start with the app that holds your money

For most businesses that means Invoicing, or Accounting if you already run a proper chart of accounts and want it inside the system from day one. This is the app that touches every client, every supplier and every tax return, so it is the one that has to be right before anything else leans on it.

Getting it right is mostly setup, not software. Load your real customer list, your real products or services with the right tax on each one, your bank accounts, and your invoice numbering the way your accountant expects it. If you sell in the UAE or Saudi Arabia, set the VAT position on every customer now, because e-invoicing rules in both countries will lean on that data, and it is far cheaper to enter it once than to clean it later.

A useful test before you move on: issue one real invoice to one real client from the system and get it paid. If that took an hour of fiddling, fix the setup. If it took two minutes, you have a foundation.

Then the app that brings the money in

Sales and CRM come next, and I would install them together because the system treats them as one flow. A lead becomes an opportunity, an opportunity becomes a quotation, a quotation becomes an order, and the order becomes the invoice you set up in step one.

The trap here is stages. People build a pipeline with nine columns because the software allows it, then nobody moves cards because nobody agrees what column six means. Start with four stages that describe what actually happens in your business, for example New, Contacted, Proposal Sent, Won. You can add more later once the team has moved a hundred deals through it and knows where the real decision points sit.

If you use outsourced callers or an appointment setter, this is also where their work lands. Every call outcome should be an activity on the lead, not a line in a spreadsheet somewhere else. When the caller and the founder look at the same pipeline, follow-up stops falling through the cracks.

Contacts and the small apps that come with them

Contacts is technically installed the moment you add anything, but treat it as a deliberate step. Merge duplicates, decide what a company record looks like versus a person record, and tag contacts by market or segment if you sell into more than one. Ten minutes on tags now saves you from a mailing list of the wrong people later.

Calendar and Discuss are also worth switching on at this stage. They cost nothing to configure, and having meetings and internal notes attached to the deal record is one of the quiet wins that makes people stay in the system instead of drifting back to WhatsApp.

Projects or Inventory, but only the one you actually run

This is where businesses split. If you deliver services, Project and Timesheets come next, because they let you see whether the work you sold is being done on time and whether it is profitable. If you sell physical goods, Inventory and Purchase come next, because stock and supplier orders are where the surprises live.

Very few small businesses should switch on both in a first build. Pick the one that matches how you make money. The other one can wait until the first is boring, and boring is the goal.

For product businesses, one warning from experience: do not import your full stock history on day one. Load current quantities on a chosen date, count them physically first, and start clean. A perfect history from three years ago is not worth the weeks it takes to reconcile.

The apps to leave off for six months

Now the part that saves the most time. These are good apps, and you will probably want some of them eventually. They just do not belong in a first build.

Manufacturing, unless manufacturing is literally the business. It depends on Inventory being clean and on bills of materials being accurate, and both take time to settle.

Website and eCommerce. Tempting because it is visible, but it pulls the project toward design debates while the invoicing foundation is still soft. Build the internal system first, then put a shop on top of it.

Helpdesk, Field Service, Quality, Fleet, and the HR family beyond basic Employees. Each one is a separate process to design, and each one asks the team to change a habit. Change one habit at a time.

Marketing Automation and Email Marketing. Without clean, tagged contacts these just send the wrong message to the wrong list faster. Earn them by finishing the Contacts step properly.

Studio and custom development. If you find yourself wanting a custom field in week one, that is usually a sign the process is not settled yet, not that the software is missing something. Run the standard app for a month, write down what actually hurts, then customize the two things that matter.

A realistic first-build sequence

Put together, a first build for a services business looks like this: Invoicing, then Sales and CRM, then Contacts cleanup with Calendar and Discuss, then Project and Timesheets. For a product business, swap the last step for Inventory and Purchase. That is five or six apps, not forty, and it is enough to run the whole cycle from first call to paid invoice inside one system.

Each step gets its own short test before the next one starts. Issue an invoice. Move a lead to Won. Find a contact by tag. Log an hour against a project, or receive a purchase order into stock. When the test passes without anyone asking how, move on.

Something you can do today

Open a blank page and write the one-sentence version of how a client goes from first contact to paid invoice in your business. If that sentence is clear, you are ready to map it onto the sequence above and you can be live on the core apps in a matter of weeks. If the sentence is muddy, that is useful too, because it tells you the work is in the process before it is in the software.

If you would like a second pair of eyes on that sentence, send it to us through the contact page and we will tell you plainly which apps it points to and in what order. No pitch, just the sequence we would run.