Credit scoring, experian & cancelling memberships?

Nine_Lives

I recently applied for a M&S card & got told today i've been refused.

It said i can check with Experian as to my details, so i did on a trial membership as it's free. So i figure i'll cancel once i've seen my score.

I got 886 & it says only 29% of people can beat this - yet i've been refused.

The problem though, is that i can't find how to cancel my membership. I go to the FAQs & there's nothing on cancelling membership, so i choose the contact us link & there's pre-set problems, but unsurprisingly none of them seem to be regarding cancelling trial membership.s

How can i get out of this so that i'm not billed?

http://www.creditexpert.co.uk

izools

You can only do it by phone

mustang1

Call them on 0800 561 0083. I cancelled mine yesterday and it only took a few minutes.

BoGoF

Oh, and your '886' was £6.95 down the drain I'm afraid.

mustang1

And don't take the refusal to heart anyway. It said my score is 999 but I got refused a credit card today too. These scores mean nothing. Each lender / card company has their own criteria they will look at eg amount of credit available and % used, income etc and if you don't fit their box at the moment, there's not much you can do about it.

izools

BoGoF wrote: »

Oh, and your '886' was £6.95 down the drain I'm afraid.

No it wasn't.

CreditExpert's new charging scheme is 30 days free, followed by £14.99 PCM, much more than it used to be.

As part of this increased fee, "Credit Scores" (random numbers) aren't charged additionally.

OP is not out of pocket.

Experian_company_representative

If you go to the frequently asked questions page and type 'how do I cancel' or even just 'cancel' the first FAQ that's returned gives you the freephone number.

On another issue, our credit scores certainly aren't random numbers. Our website explains exactly what they represent and why they're useful.

James Jones

Nine_Lives

I did go to the search & typed in cancel. I didn't see anything linking to cancellation help yesterday.

Anyway, thanks for the replies. Called up today & got a helpful woman. Cancelled no problem although we went back & forth for a little while as to why i wanted to cancel (i kept giving my answer, she kept re-asking the question).

Eventually got past that & she asked if i'd been refused credit before. I have & she pointed out my Halifax overdraft (over 4k). I wasn't aware this would be a factor. Had never asked for an overdraft, spefically asking for NO overdraft as i never use them. So that's that cancelled too now.

BoGoF

Experian_company_representative wrote: »

If you go to the frequently asked questions page and type 'how do I cancel' or even just 'cancel' the first FAQ that's returned gives you the freephone number.

On another issue, our credit scores certainly aren't random numbers. Our website explains exactly what they represent and why they're useful.

James Jones

If that's the case why are people with 'perfect' (999) scores being refused credit. I'm surprised the consumer legend that is Mr Lewis allows such 'advertising' but I've mentioned a few times there should be a MSE campaign warning against paying for a credit score and each time they've been deleted.

I wonder why......and I expect this post to vanish

savagej

The score is a number made up of the statistical analysis of the data contained in your credit report. It is a sophisticated analysis based on a balanced score card and is also a predicitive analysis weighing up factors such as percentage of credit used relative to the total amount available to you, any adverse markers on your record relative to whether or not they are paid, or being paid, and relative to how long ago they were added. Do you have a mortgage, etc..

There are many more possible analsyis that could be carried out (and are and put into the score) you can find the information on this in any good statistics text designed for a first year university undergraduate, you would also need to read something on data analysis (and mining), business intelligence and risk managment (from a mathematical view point) to help you understand what questions do you want answers to and how to go about it (there are so many questions you could ask, then when you know the question you need to understand which statistics to apply to the data, is it a non-parametric distribution, is it a Guassian distribution, how robust in the data you have, how "clean" is the data you have, how robust is the test you are applying to the data in the circumstances you are applying it and how do you test how robust your answers are?).

You would probably need a 2:1 at least in degree in a physical science or mathematics or mathematical statistics to really understand how the score are made up.

Lastly, it will futher be measured against a predictive model of populations, to get the % of other people who have this score and to give a score predictive of what a lender is likely to do with that data and that data alone.

The booklets that you get from banks on credit scoring need to be updated as now you will have one credit score based on this type of data contained within your credit file, and it will be amalgamated with data you supply about employment status, income, and time with employer etc. Then a further credit score is generated on the basis of whether or not you are deemed to be able to afford the debt, as lenders must be seen to be "responsible", then another score is calculated solely on the basis of whether or not you are considered to be over-indebted already. These two seem very similar but are in fact different.

So you might pass the general "you are deemed to be a good risk" credit score, you might be considered to be able to afford to make the repayments (they even predict, the CRA's that is, what date they think you will most likely default on your credit cards or loans), then they consider you to be over-indebted already with your other credit relative to what a responsible lender should allow someone with your income to have in total in relation to your savings with them and your other debts. There is also the issue of data validation to be looked at and how likely your application is fraudulent, this is given a percentage score the higher it is the higher the risk is consider to be, even if you are innocent you could still be declined on the basis of this with no CIFAS marker added as they did not have enough incorrect information to put one on your file, or put a flag on National Hunter, or SIRAN, or whatever the one Callcredit use (it has something to do with Geodemographics and fraud).

The area you live in also comes into it, you have person A who lives in postcode x (a very poor parliamentry consituency) and pays £2000 salary from a full time job they have had since they were 16 and are now 36 into there current account, they also have a mortgage and are very responsbile and stable and pay their debts on time. Then you have person B who lives in postcode y (a very wealthy parliamentary consituency), they are 19, never had a job and have a limited credit history say 9 months with the bank, and their benefit goes into their bank account every 2 weeks. Both are with the same bank, person A applies to upgrade from a regular current account to a fee paying account with an automatic overdraft limit and a nice shiny gold card and a credit card and gets declined, and person B does the same thing and gets a £250 automatic overdraft a shiny gold card and a credit card with £1k limit. All because people in postcode y are deemed to be of a higher socioeconomic group with superior geodemographics (as a population) and more likely to repay their debts, and make more money for the bank in the process.

As I said it is complicated.

Edit: I hope the above explains the reason why people with perfect scores from data on their files are refused credit. There is much more to this than meets the eye.

izools

Good explanantion - although you didn't touch on neural net scoring techniques like those employed by Lloyds TSB

I wonder what James Jones has to say about this