So who knows about MiData then?

Something I found out while in the UK. All banks now offer something called “MiData”. You can download the last 12 months of transactions in a standardised format suitable for loading into a comparison engine. The aim is so that the engine can look at transactions, overdrafts, fines, fees, interest rates and so on and then advise you as to whether you’ve got he best bank account deal. It’s to try and increase the churn rate of accounts, thus increasing competition.

However, how much data is there there? My own account is of very little use as my UK account hardly gets used. Couple of transactions a month and that’s it. Would anyone like to have a look at their own account? Does it pull out who a card transaction is to for example? £10 at Cafe Nero, £50 at Sainsbury’s? What level of detail does that information go to?

12 thoughts on “So who knows about MiData then?”

  1. It’s just a semi-anonymized CSV (Date, transaction type, vendor if applicable, amount, running total) – few lines from mine:

    5/10/2015, CPT, LNK RAIL NEWCASTLE CD 9732 04OCT15 , -200, 1994.76
    2/10/2015, DD, VODAFONE LTD *********-***** , -30.62, 2194.76
    1/10/2015, SO, **********, -360, 2225.38
    1/10/2015, SO, ************, -420, 2585.38
    30/09/2015, CPT, LNK TESCO Q NEWCAS CD 9732 30SEP15 , -200, 3005.38
    28/09/2015, SO, ************, -30, 3205.38
    25/09/2015, CPT, LNK TESCO Q NEWCAS CD 9732 25SEP15 , -200, 3235.38
    25/09/2015, BGC, ****************, 2435.38, 3435.38
    24/09/2015, FPO, ********************************************, -302.74, 1000
    21/09/2015, SO, ************, -30, 1302.74
    21/09/2015, CPT, LNK RAIL NEWCASTLE CD 9732 19SEP15 , -200, 1332.74
    16/09/2015, CPT, LNK 12 INTERCHANGE CD 9732 16SEP15 , -180, 1532.74
    15/09/2015, DD, RATESETTER PH******** , -50, 1712.74
    14/09/2015, DEB, NATIONAL LOTTERY I CD **** , -50, 1762.74

    CPT – Cashpoint
    DD – Direct debit
    SO – Standing order
    BCG – BACS payment
    FPO – Payment to 3rd party

  2. People are sufficiently fucked up to think uploading their bank transactions to a “comparison engine” is a good idea?

    Can’t people just add up their own banking charges and look around for a better deal if they think it’s too expensive?

    Oh, wait…

  3. Why would one share one’s bank data? It’s worrying enough, thinking about what banks are capable of doing with data. And then accessing that data depository through a smartphone. All of which leak like sieves, due to the apps they’re infested with. (Clue: That they’re “free” doesn’t mean they’re free.)
    Why not just publish your financial records on a free advertising site? Save the hacking community* all the bother of writing code.

    *Not to mention the Surveillance State. if there is a distinction.

  4. The MiData shows which Nero/Sainsbury/cash machine the transaction took place.

    Given that banks, supermarkets and marketeers are already datamining ours transactions with them for their own benefit, a comparison would have to really share the benefits of datamining with uploaders to make it worth their while. Then again, young people today don’t care much about privacy, so it wouldn’t be take very much.

    (Not normally) Anon earns around £37,000 a year, carries out most of his day to day spending in cash. The day before pay day he puts any balance above £1000 into a savings account. There are two standing orders at the start of the month, presumably mortgage and bills. There are no petrol costs so he isn’t a driver. The mortgage seems a little low to be a sole occupier, so I think he goes halves with a wife. He lives in the centre of Newcastle, somewhere between the rail station and the Quayside Tesco Express.

  5. Commutes into the centre of Newcastle I imagine. Two ex-wives, one sprog is clued-up enough to get his/her pocket money by standing order.

  6. Given the number, frequency and amount of the cashpoint withdrawals of (not normally) Anon, my conclusion is that the withdrawee is a regular user of cheap providers of sexual services in the vicinity of Newcastle Railway Station – probably hand and blow jobs.

  7. The point of Midata appears to be: Make it a doddle for consumers to get their own info with the hope that many of them then give it away to comparison services and suchlike. The pushers of Midata even expect councils to be getting in on it.

    Here is a briefing pack (pdf) written by the Department for Business, Innovation and Skillz. It sets out some of the principles behind the project and provides examples of how it should work.

    I find the tone of it a little unsettling.

  8. 15/01/2015 VIS SAINSBURYS PETROL CARDIFF-£41.90 -£1,053.51
    15/01/2015 VIS THE TYRE SHOP SWANSEA* * -£77.94 -£1,011.61
    15/01/2015 ATM CASH RB SCOT JAN15TESCO [email protected]:19 -£10.00 -£933.67
    16/01/2015 VIS SAINSBURYS S/MKTS CARDIFF **** -£5.25 -£1,001.16
    16/01/2015 VIS POST OFFICE COUNTECARDIFF -£16.40 -£995.91

    Curiously, my MiData has different transaction type abbreviations, and I get the time that cashpoints were used.

  9. Barclays doesn’t seem to have a ‘midata’ button as such. It does have a download capability in CSV and several accounting package formats. The CSV of my account looks nothing like the midata samples above. The fields are Number (blank in my sample), Account (sortcode/acct no), Amount, Subcategory, & Memo. This last field is basically three fixed size fields concatenated – Payee, a reference, and the transaction type (STO, BCC, etc) as the last three characters.
    I assume midata ought to conform to a standard layout so the Barclays stuff ain’t it.

Leave a Reply

Your email address will not be published. Required fields are marked *