npaskin
New Member
- Messages
- 506
To be fair to How Many Left, they get their data from the Dept of Transport (DVLA database), and they do say on their site: “As with all large government databases, there are errors in this dataset (especially since quite a lot of the data for older vehicles is based on paper records that were originally maintained by local authorities). The most common error that crops up is vehicles that don't have exactly the correct model variant recorded on their V5 registration document…the key to understanding whether or not the data is accurate for your particular model is to check the model name on its V5 registration document. If it's not what you expect it to be, then it's likely that the DVLA statistics for that model aren't very accurateâ€.
For actual sales of a marque, the SMMT data is going to be better simply because it reflects actual money changing hands and it is at general marque level (but still likely to have problems re timing of pre registration sales, special discounts, press cars, etc.). You get an idea of how problematic the DVLA data is if you search e.g Quattroporte and you get 13 different names, some of which are very specific and others just general – I imagine the average input clerk has no clue about exotics like Maserati, and dealers doing registration docs have no way to be consistent in their classification - both classic reasons for rubbish datasets. As long as the same set is used for comparison it can be useful, eg the SMMT showing a 20% decline in year on year UK Maserati sales is likely to be pretty accurate I would think.
For actual sales of a marque, the SMMT data is going to be better simply because it reflects actual money changing hands and it is at general marque level (but still likely to have problems re timing of pre registration sales, special discounts, press cars, etc.). You get an idea of how problematic the DVLA data is if you search e.g Quattroporte and you get 13 different names, some of which are very specific and others just general – I imagine the average input clerk has no clue about exotics like Maserati, and dealers doing registration docs have no way to be consistent in their classification - both classic reasons for rubbish datasets. As long as the same set is used for comparison it can be useful, eg the SMMT showing a 20% decline in year on year UK Maserati sales is likely to be pretty accurate I would think.