i i'm professor marketing american university change in the night or inference

paul williams i recently categorical probably children's services marketing that builds on some research we

conducted over the last are yours this data flow from the time to try and

we have used for about eleven years prior to getting that in domains

by biased in human a large size firms forty one for and we would want

to studies with these four we present to secure exact in organisations you realise there's

millions of dollars we made by a single by one or two or three percentage

points

and so that focus was their initial focus only in customers the pointing it performs

organisations in two thousand five we will consider your marketing and she solicited articles from

somebody scholars and you'll just real pretty webster don't limit

and others ask them based on your what with the shortcomings what you're the consists

of those articles in that by j was that we in the morning more

may actually work we show a little bit of this morning actions and decisions on

financial performance in for specifically we need to be able to document the financial performance

metrics relating to read from increasing task force that for changes in product or service

delivery

talking back to the financial performance of the four original

so what we came up with it will be the stores to justify was that

what we try to previous years

at that i don't decided using so these longitudinal databases for the research so that

way well as i started analysing the data to see what happens changes over time

that strategies that's the final performance

we focus is that the improvement over most studies are done feel because most a

cross sectional those of eight three hundred by either way time but it doesn't show

that i am i change over time and l s played by the performance

so what or restaurant we went back to looking at something spanish for metrics be

variables look row rescore share a

r y are redraw tested so you try to find estimates

what we try to do it stays databases strategies specifically customer satisfaction representations attractive over

a period of three four five years so that are inside to okay is that

into the changes by four

so that i like to earn representation over all and he saw that some of

the specific financial metrics that we the results came out

okay and you will a particular study we were dealing with a building services industry

when we talk about some heating ventilation and air conditioning systems most exciting research are

want to take relevant and we're dealing with what complex onsets financially as well

there were three main results we found the data that the first one is actually

very strong relationship between customer satisfaction and actual attention with the for the was a

renewal right wherever the satisfaction models

okay but most there is just going to be the case now we got ninety

hours of data group

the second interesting result

was that was for you growth particularly it was a very strong relationship between customer

satisfaction hundred and you wrote cable leading to high level of satisfaction lead to higher

levels of running you wrote with respect to the customer

another interesting statistic we found in our data was customer satisfaction is highly correlated with

small performance is represented by stock price and we also confirm that tokens q six

actually correlation between the model performance for low prices

and what the customers are satisfied

conclusion a i think we show longitudinal data has been extremely insightful interesting and we

have confirmed favourites already

shown in conceptual study we think it is i value to use a more accurate