What software an Israeli SMB actually runs, why the gaps between those systems are where the work sits, and the four seams that produce almost every manual process a business complains about.
Key takeaways
- Israeli businesses rarely lack software. They have five or six systems that each work well and do not talk to each other, and the manual work lives in the gaps.
- The invoicing system is the one nobody replaces. It serves the accountant, not the business, which is why integrations should adapt to it rather than the reverse.
- WhatsApp is the primary customer channel in Israel and is almost never connected to anything. That single gap produces more lost leads than any other.
- Count the seams, not the systems. Four or five connection points is normal; each one is either automated or is somebody's manual routine.
When an Israeli business says "we need a system", the answer is almost always that they already have five. They simply do not talk to each other, and in every gap between two of them sits a person copying data.
This article maps the typical stack and its seams - because that, rather than inside the systems themselves, is where nearly all the manual work a business complains about actually lives.
What is actually running
| Layer | Typically running | Who really owns it |
|---|---|---|
| Invoicing and bookkeeping | Morning, iCount, EZcount, Rivhit, Hashavshevet | The accountant |
| ERP (mid-size and up) | Priority, SAP Business One, Comax | Operations |
| CRM / work management | Monday, Fireberry, HubSpot, Zoho - and frequently Excel | Sales |
| Payments | Cardcom, Tranzila, PayPlus, Meshulam | Finance |
| Store (where relevant) | Shopify, WooCommerce, Wix | Marketing |
| Customer communication | WhatsApp, email, sometimes SMS | Everyone, meaning nobody |
Two notes on the table:
The ownership column is the important one. Every system has a department that holds it, so every integration has two stakeholders - which is why such projects stall on decisions rather than on code.
Excel is part of the stack. There is almost always at least one critical spreadsheet someone maintains by hand and nobody mentions in the requirements. Ask about it explicitly.
The system nobody replaces
The invoicing system is a fixed point. It serves the accountant who works in it daily, not the developer who integrates with it once.
The practical consequence: recommending a change of invoicing system because its API is nicer is bad advice. The cost is not technical - it is migrating history, retraining, and risk during the transition. The integration adapts to the system, not the reverse. Detail in the Israeli invoicing API comparison.
The four seams - and where the work is
Seam 1: lead → CRM
A lead arrives from a website form, a phone call, WhatsApp or a campaign. Someone types it into the CRM.
What breaks: leads are lost outside working hours, and the lead source is not preserved - so you cannot tell which marketing channel works. That sounds minor and is not: without source, the entire advertising budget is guesswork.
Seam 2: deal won → ERP or invoicing
The sale closes in the CRM. Now a customer, an order and a document are needed in the operational system.
What breaks: duplicate customer cards. This is the most expensive and quietest failure in the whole stack - it fails no test, it accumulates, and it surfaces six months later in a report that does not reconcile. Detail in syncing monday with Priority.
Seam 3: payment → invoice
The customer paid through the gateway. A tax document is needed.
What breaks: double issuance. If the gateway is configured to issue a document and the system issues one, there are two tax documents for one transaction. This is the most common design mistake when payments and accounting touch the same project, and it surfaces at month end.
Seam 4: status → customer
The order shipped, the invoice was issued, the meeting is booked. The customer wants to know.
What breaks: nobody updates them, so the customer calls to ask. Every such call is a cost that an automated message would have prevented.
The biggest gap: WhatsApp
In Israel, WhatsApp is the primary customer communication channel - and in most businesses it is completely disconnected from the rest of the stack.
Which means: sales conversations happen in a channel with no record, no assignment to a rep, and no follow-up. A lead arriving on WhatsApp on Saturday and unanswered until Sunday afternoon is a lost lead - and nobody will know it existed, because it never reached any system.
This is the one gap where the distance between "what is technically possible" and "what actually happens" is largest. The WhatsApp Cloud API exists, it is official, and it solves this - but it requires a setup most businesses do not know is available.
How to map a stack in practice
Before recommending anything, four questions. I ask them in the first meeting and they are almost always enough:
- "Describe what happens from the moment a lead arrives until the money is in the bank." - every place the answer includes "and then I copy" is a seam.
- "Which spreadsheet is critical to you?" - the answer is always yes, and it always reveals something that was not in the requirements.
- "What happens when that person is on holiday?" - exposes processes dependent on one individual.
- "When did the reports last fail to reconcile, and why?" - points directly at the broken seam.
What to fix first
Not everything at once. The priority order I work by:
| Priority | Why |
|---|---|
| 1. The gap losing leads | Easiest to quantify and most persuasive - a lost lead is revenue that did not arrive |
| 2. The gap creating duplicates | The damage compounds and surfaces late, making it more urgent than it looks |
| 3. The recurring manual gap | Priceable: how many hours per month |
| 4. Everything else | After the first three are stable |
What not to do: start by replacing a system. A business that changes its ERP to solve an integration problem buys a six-month project instead of a one-month one. Connect what exists first; if a problem remains after that, it is a different problem and can be discussed separately.
For those reaching the actual connection stage: connecting an Israeli ERP to anything. And for those who discover a system has no interface: my software has no API.
Frequently asked questions
What software does a typical Israeli business run?
Usually an invoicing system (Morning, iCount, EZcount, Rivhit or Hashavshevet), a CRM or work-management tool (Monday, Fireberry, HubSpot, Zoho - often Excel), a payment gateway (Cardcom, Tranzila, PayPlus, Meshulam), sometimes an ERP such as Priority or SAP Business One, and WhatsApp as the main customer channel. The problem is rarely a missing system - it is that none of them are connected.
Where does the manual work in a business actually come from?
From the seams between systems, not from inside them. There are typically four: lead into the CRM, deal won into the ERP or invoicing, payment into an invoice, and status back out to the customer. Each seam is either automated or is somebody's manual routine, and counting seams rather than systems is what tells you the size of the work.
Why is WhatsApp the biggest gap in an Israeli business stack?
Because it is the primary customer channel in Israel and is almost never connected to anything else. Sales conversations run in a channel with no record, no assignment to a rep and no follow-up, so a lead arriving on a Saturday and unanswered until Sunday is lost - and nobody knows it existed, because it never reached a system. The WhatsApp Cloud API solves this, but most businesses do not know it exists.
Should a business replace a system to fix an integration problem?
Almost never as a first move. Replacing an ERP or invoicing system to solve a connection problem turns a one-month project into a six-month one, and adds history migration, retraining and transition risk. Connect what already exists first; if a problem genuinely remains afterwards, it is a different problem and can be evaluated on its own terms.
What questions reveal where a business is losing time?
Four are usually enough: describe what happens from lead arrival to money in the bank (every "and then I copy" is a seam); which spreadsheet is critical to you (the answer always reveals something absent from the requirements); what happens when that person is on holiday (exposes single-person dependencies); and when did the reports last fail to reconcile and why (points directly at the broken seam).
Keep reading
Related service
Custom CRM
A CRM built around your pipeline, connected to the tools you already use.
About the author
Yehonatan Saadia
Freelance automation, web & MVP engineer
I'm Yehonatan Saadia, a senior engineer who builds business automation, custom websites, and MVPs for small and mid-sized companies across the US, Europe, and Israel. These guides come from real client work, not theory.
Work with meHave a project like this?
Tell me what you're trying to automate or build and I'll tell you the fastest reliable way to ship it.