You must be logged in to reply to this topic.
- MA2572 applied to small sales business
-
Currently I am working for DSII – a small company selling voting systems with customized software. Essentially in sales & support department. Probability and Statistics are important in all sales related businesses. One aspect of statistics that we deal within DSII specifically is inventory and projections. The way our supplier works is that they need 3 months from the date of order, to date of delivery. The way our sales work is that they are shipped out within 24 hours after confirmation – and sales can happen any time. To add in a few more variables, we have customers who will rent, small or large quantity of units, for a short or long amount of time, and customers who will do a permanent purchase, once again small or large quantity. Plus, we have “business seasons” as well – where certain months will see higher levels of business and others will be quiet months. Essentially these variables get factored into inventory projections. How much of a certain model do we keep in stock? When do we place the order? I think just this small scenario that I described is exactly what Probability and Statistics is made for. Obviously with any businesses – focus is maximizing profit. As I continue with MA2572, I would be interested to see how the different models can be applied to see a better picture of DSII. Relationship between expenses in advertising and actual sales – or at the least inquiries (leads). One thing that I do in the company is provide online demo meetings for our product and its applications (relating to the customer’s specific event/needs). So, I would like to see how effective that is in confirming their purchases. Relationship between doing demos and getting the sale.
I applied Hypothesis Testing to see if a demo influences customer’s purchase with our company. I took the data from Oct 1st to Oct 31st. Had a total of 40 customer leads. 22/40 demo were done, and 13/40 total purchase. Initial hypothesis being “Demo has NO effect on purchase”
Total 40 Demo D(n) D(y)
Purchased Purchased 0 1
1 Yes 13 P(n) 0 15 12 27
0 No 27 P(y) 1 3 10 13
Demo 18 22 40
1 Yes 22
0 No 18
Expected 0 1
0 12.15 14.85 27
1 5.85 7.15 13
18 22 40(O-E)^2/E 0 1
0 0.668 0.546
1 1.388 1.136
3.73996374 TS
1 DOF
0.053125598 P-ValueSince the resulting P value is 0.053, grater then 0.05 — Initial Hypothesis stands. Demo has NO EFFECT on purchase decision.
The sample size is very small with just 1 month, and also the data is not 100% accurate. For example, one customer might ask for quotes on a system for 2-3 different models. Where we will do a demo for all 3 systems (in 1 session or multiple sessions). But they will purchase just 1 system. The other models in this data gets counted as a “not purchased.” With the resulting P-value being so close – even with the limited data – i am inclined to believe that continuing to do Demo before purchase with a customer is still beneficial.
You must be logged in to reply to this topic.
