On 28 May 2008, at 19:12, ..hidden.. wrote:
Quoting Stroller <..hidden..>:Yes, but there's an obvious difference between: 00042 00043 00044 00046 (missing number precedes) and: 20080501 20080502 20080503 20080512 (issued on the last day of May) 20080601 ... 20090101 If one accepts that date-based invoice numbers are acceptable (and I contend the taxman might do so), no-one expects to see invoice number 20081301, and the "gap" between 20080512 and 20080601 is quite logical & reasonable. It's the anomalies that raise the taxman's hackles, not the choice of implementation.Erm, you mean the "gap" between 20080531 and 20080601, surely?
No. There are only 12 invoices issued in May in this example.Presumable the issuer is a consultant, or someone who otherwise only works for a handful of companies at day rates.
Tch! This should be obvious!
'Cos Iposit that 20080525 is a perfectly reasonable number in this scheme. ;)
Yes, but that number hasn't been used by the issuer.
This scheme, in any case, would only allow one invoice per day.
No, it allows at least 99 per month. Only the year and month are date- based. The following digits are incremental. So in a busy month 20080687 might be issued on the 25th. In fact, there's nothing to stop the issuer going on to 200806100, 200806101 if he needs to.
Admittedly, my example is quite unclear. The invoices I've seen with date-based numbering have always been in the 8-05/01, 8-05/02, 8-05/03 format. Nevertheless, it is not clear to me that the skip to 8-06/01 strictly meets the "consecutive" requirement you posit.
I'll buy a professional opinion from my accountant when next I see him.
Always advisable.This is an academic discussion because, as I said before, I use fully incremental numbers, too.
Stroller.