# 17. Recommendations to the Client

We will deliver our findings using a series of charts. We will then recommend a series of next steps when implementing the overbooking policy.

**Insights from the model**

- Additional profit of £850-940,000 if overbooking policy had been in place

- Additional profit is sensitive to voucher costs but only at very high booking limits

- 7 flights provide over 90% of the profits, one flight provides a loss

**Recommendations to client**

- Introduce overbooking policy on most profitable routes first

- Test various overbooking policy levels on each flight

- Replace our model input assumptions with real data over time

As you recall from earlier lessons our problem statement states that if we had implemented an overbooking policy with a common booking limit across some or all of our flights, by how much would we have improved profitability without adjusting prices in the last three months? Our first chart shows total additional profit for a range of different booking limits and conversion rates. If we had set our booking limit to 200 seats we estimate that an additional profit of 850 to 940,000 pounds would have been achieved over the three month period. This assumes a voucher costs of 170 pounds per bumped passenger. 5 00:00:45.17 --> 00:00:58.26 For varying booking limits, the total additional profit exhibits and N-shaped curve. Below a certain level you leave money on the table by underestimating demand and keeping the overbooking policy too low.

However, beyond a certain point, too many tickets are sold and the number of bumped passengers increases reducing profitability. Next, let's take a look at voucher costs. The voucher cost assumption is critical because the company is still unsure how much money is needed to compensate bumped customers. As you can see from this chart, a higher voucher cost does impact profitability but the company will still make a substantial additional profit if the booking limit is set correctly. Although losses at higher booking limits look troubling, in reality, losses of this nature are unlikely to occur as the company will be able to tweak the booking limit over time and fine tune the policy somewhere in the profitable range. Our next insight is with regard to the value created for each flight. 7 flights provide over 95% percent of the profits. With one flight, London to Berlin at 9:15 exhibiting a loss of 17,000 Obviously, the company's best advised to introduce an overbooking policy in the more profitable flights first. For the flights that exhibit no gains or losses, they were simply never full, so extra tickets couldn't be sold. To create this chart in Excel, I simply sort it, total additional profit, from largest to smallest, then select the data. Alt + N, C to create a column chart and off camera I perform some formatting to make this chart look like it appeared on my PowerPoint slide. So where does this leave us in terms of recommendations? Although our model can't tell us how much additional profit and overbooking policy we'll make in the future, it's reasonable to assume that Zippy is likely to make significant additional profits on some routes if the policy is implemented. We would recommend to Zippy to implement the overbooking policy initially on the most profitable routes. We would also suggest when implementing the overbooking policy to test various policy levels initially rather than picking one and sticking with this for a three month period. And lastly, we would gradually refine the model with real data for voucher costs, conversion rates, and demand, replacing the assumptions that we made from the first three months of the year. Models by their nature are simplified and what we have created in this course is no exception. There are a number of obvious limitations to this model which should also be highlighted to the client. I'll explain these limitations in detail in the next lesson.

# Contents

#### 1. Introduction to Modelling

2 mins

#### 2. Framing the Problem

3 mins

#### 3. Diagram the Problem

3 mins

#### 5. Lay Out Model in Excel

3 mins

#### 6. Build Simplified Model

4 mins

#### 7. Sensitivity Analysis

4 mins

#### 9. Estimating Market Demand

5 mins

#### 10. Complex Model formulas

5 mins

#### 11. Perform Checks on Model

4 mins

#### 12. Calculate Model Outputs

3 mins

#### 14. Goalseek and Solver

4 mins

#### 18. Limitations of the Model

2 mins