Every mid-sized company in Jordan we speak to has a digital transformation initiative. Most are stalled. The reason is almost never the technology choice — it is that the work was sequenced in the wrong order.

This is the sequence we use, and why each step comes where it does.

Step one: find out what actually happens

Before choosing any system, document how work really flows — including the informal parts. The spreadsheet someone maintains privately, the WhatsApp group where approvals actually happen, the report that gets rebuilt by hand every month because nobody trusts the automated one.

These are not inefficiencies to be embarrassed about. They are the system compensating for a gap, and each one tells you precisely where the real requirement is. Automate the official process while ignoring the shadow process and you will build something nobody uses.

Budget two to four weeks. Talk to the people doing the work, not only their managers.

Step two: fix the data before the applications

This is the step most programmes skip, and skipping it is why they stall in month five.

If your customer list exists in four places with different spellings, no CRM will fix that — it will inherit the mess and add a licence fee. If your product codes are inconsistent, your new ERP will report inconsistent numbers faster.

Data cleanup is unglamorous and nobody wants to fund it. It is also the highest-leverage work in the entire programme.

Step three: pick the smallest valuable thing

Digital transformation in Jordan tends to be scoped as a two-year, everything-at-once programme. Those fail, because by the time anything ships the business has changed and the sponsors have moved on.

Instead, pick the single process that costs the most in wasted hours and fix that first. Ship it in eight to twelve weeks. Then use the credibility from that delivery to fund the next one.

Sequencing by value rather than by system architecture also means that if the programme is cut short, you keep what shipped.

Step four: plan for the adoption dip

Productivity falls when a new system launches. Always. The question is how deep and how long, and both are functions of preparation rather than software quality.

  • Train before go-live, not during.
  • Keep the old system readable — not writable — for a defined period.
  • Identify a champion in each team who learns first and helps their colleagues.
  • Expect four to eight weeks before productivity returns to baseline, and communicate that upward so nobody panics in week two.

Realistic budgets

For a Jordanian company of 50 to 200 staff, a first meaningful phase of digital transformation typically lands between 30,000 and 120,000 JOD across twelve to eighteen months, depending on how much bespoke integration is required.

The split is usually a third on software, a third on implementation and integration, and a third on change management and training. Programmes that under-budget the last third are the ones that stall.

What to be sceptical about

  • Any vendor who quotes before understanding your process. They are selling licences, not outcomes.
  • Big-bang launches. If everything goes live at once, everything fails at once.
  • "AI-powered" as a headline feature without a specific decision it improves. Ask what it predicts and what the current accuracy of that prediction is.
  • Roadmaps beyond eighteen months. Nobody can see that far. Plan the next two phases in detail and sketch the rest.
The companies that succeed at digital transformation in Jordan are not the ones with the biggest budgets. They are the ones that shipped something small, proved it worked, and used that to earn the next round.

Measuring it

Pick metrics that a sceptical CFO would accept:

  • Hours per week returned to staff, measured before and after.
  • Error rates in the specific process you changed.
  • Cycle time from request to completion.
  • The cost of the errors you stopped making, which is usually larger than the labour saving and almost always uncounted.

Where to start this week

Ask three people in different departments to describe how one shared process works. If you get three different answers, you have found your first project — and you did not need a consultant to find it.