I go for NSC because I don't use a lot. SC can be a significant proportion and the other scheme of high initial unit cost was v. bad for me as most of my useage was in that bracket. Every time I try a site the savings seem to be negative (I'm with Ebico atm) but perhaps it's time to check again.
They were. I noted it, and that I would have to plan an exit strategy carefully.
I am not really expecting to be with them that long. I do my readings regularly anyway.
That's what I'm on...and I have a note in the calendar.
I am prepared for that. I project an annual saving of £900 approx., so it's worth it!
Irrelevent, unless you are expecting some absolute measure of performance from online reviews!
I cannot see any reason why a different proportion of customers would submit reviews for a large company vs. a smaller one. Therefore a relative rating is useful.
If what you say is true then all companies would get the same (i.e. 0) customer satisfaction score.
In article , Mark scribeth thus
No not at all, it's just 'uman nature. People are far more prone to grumble and complain than they are to praise.
Therefore their far more likely to complain than to comment on good service..
I know but it's not relevant.
In article , Mark scribeth thus
Gawd.. Its very relevant, its simple people are far more likely to complain about a fault or a problem or bad service than otherwise!.
Don't you?, SWMBO here does it all the time so its got to be right;!...
In message , tony sayer writes
Are you not brave enough to say "'er in the kitchen" ?
In message , Doctor Drivel writes
Pound coins, adam
I was with E.on for a while and didn't have any problems with them (I only changed provider because I moved house).
I'm currently in the process of changing to Npower and, like you, I was a little worried about the reviews but decided to try it as the savings are quite significant.
-
Oh Dear. I will explain it with a simple example:
Company A: Has 1,000 customers. 100 reviews, 50 bad and 50 good. Overall review: 5/10
Company B: Has 10,000 customers, 1000 reviews, 800 bad and 200 good. Overall review: 2/10.
Therefore we can conclude that company A is "better".
The fact that more people are likely to review a company if they have a bad experience is unimportant since this factor can be cancelled out when making the comparison.
Simple.
Fantastic Maxie. Who is this pounds, coins adam company?
Well guv!, I know my place and its not in there. Shes from France and French wimmen tend, in general, to treat that place as their hallowed ground and domain. As she puts it your fine out there in that workshop and office of yours, but please don't transgresses in here, this is my Manor;!...
And bl^^dy good she is in it here to thats one good thing about having an Imported missus they Know -how- to cook proper like;)...
No, your still missing the point re your fundamental data here, and thats that people are far more likely to go to the trouble to complain than to praise..
good.
But that would only skew *all* ratings *equally* towards the bad. This "lack of praise" doesn't matter as the "rating" isn't real numbers but a ratio of good v bad.
In message , tony sayer writes
"Elle dans la cuisine", then
I have a "dia dalam dapur"
hot and spicy ...
Right last time;!..
We have 100 people who are supplied by a service from a company.
Half of them are very pleased with the product.
Half of them are most displeased.
So that should be a satisfaction rate of 50%
Now thats if everyone who got this service or disservice would comment or complain Positively and Negatively.
Now I put to you that more will by nature complain than praise.
Thus distorting the real picture....
The values may not be "true" ie your 50% above may come out as 30% but that isn't relevant as all the results from all companies will be skewed by the same amount.
So a company with a 20:80 bad:good "true" ratio may only come out at
50% in actual submitted bad:good reviews but that figure is still "better" than the other company with the similary skewed 30% figure.
Why?..
If the data isn't there in the first place then how can your results be accurate ?..
Because if you don't assume the skew, due to happy people being less likely to make that known, is not the same across all companies it implies that you can't base any comparison between companies on the numbers of bad or good reviews in any way what so ever. Personally I feel that provided the samples for each company are reasonably large the skew will be the same across companies(*).
We aren't talking about accuracy or real numbers in the rating only a probability that Company A is better, FSVO "better", than Company B.
(*) Assuming that none of the companies have been the target of a hate, or love, campagne that is.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required