What drives the timeline
| Your situation | Realistic timeline | The long pole |
|---|---|---|
| Modern ERP, clean data, standard invoicing | 6-8 weeks | Provider integration and testing |
| Established accounting system, some data gaps | 2-3 months | Customer record clean-up |
| Mixed systems, complex billing (retention, milestones) | 3-4 months | Mapping billing model to compliant format |
| Spreadsheet billing, inconsistent data | 4-6 months | Rebuilding the invoice data foundation |
| Multi-entity group, shared customers | 4-6 months | Consistency and intra-group transactions |
Notice that in only the first row is the software the constraint. In every other row the long pole is data or process, which is exactly the part a provider cannot do for you and the reason starting early matters.
Why that is the answer
It helps to see implementation as four phases that partly overlap rather than one project with a single duration.
Assessment (1-3 weeks). Establish which revenue band you are in, audit your customer and product data against the required fields, and map how each invoice type you issue (standard, credit note, retention, advance) must look in the structured format. This phase produces the honest timeline for everything after it.
Provider selection and contracting (2-4 weeks). Shortlist accredited providers with a native connector for your system, test them against your edge cases, and contract. This can run in parallel with data clean-up.
Integration and data remediation (3-12 weeks). The variable phase. Connecting the provider is quick; correcting customer records, categorising line items for tax, and adjusting the accounting workflow so it produces compliant data is where the months go.
Testing and go-live (2-4 weeks). Exchange real invoices in the voluntary pilot from Open from 1 July 2026, fix what breaks, then cut over. Testing against real transactions rather than samples is what stops go-live becoming the first time you discover a problem.
Why data, not software, sets the clock
The structured format demands fields many businesses do not currently capture cleanly. Every customer needs a valid tax registration number recorded against them. Every line item needs a tax category. Units of measure need standard codes. None of that is difficult in principle, but at scale it is a data project, and its length depends entirely on how consistent your records already are.
A business with 200 customers and clean records might close the gaps in a fortnight. A business with 5,000 customers, duplicates, missing registration numbers and free-text product descriptions could spend two months on the same task. The provider’s software is identical in both cases; the timeline is not, and the difference is invisible until someone actually audits the data.
This is why any credible estimate starts with a look at your records rather than a look at your revenue.
The billing models that add time
Standard invoicing maps to the structured format easily. Certain billing patterns do not, and each one adds weeks because the mapping has to be designed rather than assumed:
Retention billing in construction, where a portion of each invoice is withheld and released later, has to be represented so both the withholding and the release produce valid invoices. Milestone and progress billing needs each stage to map to a compliant document. Advance payments and deposits held outside the accounting system are a common trap, they generate a tax point that the e-invoicing flow must capture. Subscription and recurring billing needs the recurring engine to feed the provider cleanly. Intra-group transactions have a transition running through 1 January 2029, which affects timing for groups.
If any of these describe how you bill, add time and start earlier. They are the edge cases where implementations run over, precisely because they were treated as afterthoughts.
How to compress it safely
You cannot compress the data work by wishing, but you can sequence the project so the deadline is met even if not everything is perfect on day one.
Start the assessment immediately. It is cheap and it produces the real timeline. Run provider selection in parallel with data clean-up rather than in sequence. Prioritise your largest customers, since they are the ones who will reject non-compliant invoices first, so getting them compliant protects the most cash soonest. And use the voluntary pilot from Open from 1 July 2026 to test in a live environment, which turns go-live from a leap into a formality.
What does not work is buying the software first and discovering the data problem second. That sequence guarantees the integration stalls waiting for records that should have been cleaned in parallel. The businesses that go live smoothly are the ones that treated it as a data project with a software component, not a software purchase with a data footnote.
The common misunderstanding
- Estimating from revenue rather than data quality, when data quality is what sets the timeline.
- Buying the provider first and finding the data problem second, which stalls the integration.
- Treating retention, milestone and advance billing as afterthoughts, when they are the edge cases that run over.
- Testing on samples rather than real transactions in the pilot, so problems surface at go-live.
- Running phases strictly in sequence when selection and data clean-up can overlap.
- Assuming a short project because the software connects quickly: connection is not readiness.
- Starting late on the theory the deadline will move, when data work has its own irreducible pace.
What to do next
- Run the assessment first: it produces your real timeline in one to three weeks.
- Audit customer records for missing tax registration numbers, the most common delay.
- Map your non-standard billing: retention, milestones, advances, to the compliant format early.
- Select a provider in parallel with data clean-up, not after it.
- Book time in the voluntary pilot from Open from 1 July 2026 to test before go-live.
Related questions
Frequently Asked Questions
How long does e-invoicing implementation take?
Six weeks to six months. Clean data on a modern system: six to eight weeks. Spreadsheet billing with inconsistent records: four to six months. The software connects quickly in every case. The range comes from the state of your data, not the provider.
Why does it take months if the software just connects?
Because the requirement is structured invoice data, and most businesses do not capture every required field cleanly today. Adding valid tax registration numbers to every customer, categorising line items for tax, and standardising units is a data project whose length depends on how messy your records are.
When should we start?
Now, for most businesses under AED 50 million. Go-live is 1 July 2027 with a provider appointed by 31 March 2027, and working backwards through data clean-up, integration and testing consumes the intervening months for any business with data gaps.
Which billing models take longer?
Retention billing, milestone and progress billing, advance payments held outside the system, and subscription billing all add time because each has to be mapped to a compliant invoice rather than assumed. If you use any of them, start earlier.
Can we go live in a few weeks if we have to?
Only if your data is already clean and your billing is simple. You can appoint a provider and open the channel quickly, but true readiness, compliant invoices for all customers and billing types, cannot be forced faster than the data work allows.
Does the voluntary pilot shorten the timeline?
It does not shorten the data work, but it de-risks go-live by letting you test against real transactions from Open from 1 July 2026. Businesses that use the pilot cut over smoothly; businesses that test for the first time at go-live discover problems at the worst moment.
Can the provider speed it up?
The provider can make integration efficient, but it cannot clean data it never sees or redesign your billing workflow. The parts that determine the timeline are on your side of the boundary, which is why an early assessment matters more than provider choice for scheduling.
What if we run multiple entities?
Plan for the longer end, four to six months. Multi-entity groups have to be consistent across companies, often share customers whose data must reconcile, and have intra-group transactions with a transition running through 1 January 2029. The coordination itself adds time.
What is the first thing to do?
A readiness assessment: confirm your band, audit your data against the required fields, and map your invoice types to the compliant format. It costs little, takes one to three weeks, and turns a vague deadline into a dated project plan.
Tell us your accounting system, customer count and how you bill. We will assess your data against the required fields and give you a realistic timeline to 1 July 2027, not a generic one.
Check my compliance status 058 101 9570
Last reviewed 27 July 2026. Rates, thresholds and deadlines change, the e-invoicing provider deadline has already moved once. Confirm current requirements with the Federal Tax Authority before acting, or ask us to check your position.