Modeling sales and profit can be done in different ways, depending on your nes. You can model them quickly with a certain degree of error, or in a more reliable and accurate way by spending more time. If you have limit resources, a quick and simple method is better than nothing. A quick estimate can be useful, for example, for a B2B business, where you can simply estimate the efficiency and costs of your sales managers, summarizing them. The following data is an example of what data you ne to be able to model sales and profit:
- How many leads are generat per manager per unit of time;
- How many phone calls per sales manager per unit of time;
- How many meetings per manager per unit of time;
- How many sales are there per manager per unit of time;
- Estimate the cost of acquisition (e.g. marketing budget over time).
A similar super-fast and simple method can be us for online business, taking into account the following data:
- Estimate the cost of your telemarketing data marketing budget over time, for example, ₽50,000
- Estimate the average cost per click, for example, 10 ₽ => 5,000 clicks
- Estimate the conversion rate, for example 5% => 250 sales
- Estimate the average basket check, for example, 1000 ₽ => 250,000 ₽
- This gives you a monthly profit forecast of ₽200,000.
Once you have creat an estimate like the one above, be sure to double-check it from top to bottom; estimate your market share and see how realistic it is. If you have a limit sales throughput, say 200 sales per month, you won’t be able to have more than 200 sales, even if your model shows 250 sales per month.
Capacity utilization >90% is usually this technology will only guarantee a competitive edge quite high for any business (unless you are heavily investing in automation and scaling services at the level of a company like Amazon). If you are achieving this high level on a regular basis, congratulations on your good results. In some cases, you may also consider seasonality, for example, Christmas decorations are mostly sold before the New Year, not in the summer. Quick and easy methods like the ones describ above are all you can apply to start a profitable business in your field.
How to Model Sales and Profit (A More Accurate Way)
When you have more free time and taiwan data resources, you will be able to apply a more reliable and accurate modeling system. To make such modeling possible, you ne to collect and systematize the following data:
- Data on units sold for each article (warehouse inventory unit) broken down by date;
- Potential interest in each product over a certain period of time (for example, tracking visitors using Metrica, Google Analytics and other data analysis tools );
- Supplier data by article number;
- Sales geography for each item;
- Data on past promotions for each item;
- Customer feback data for each item
- Data on these items from competitors (prices, discounts, their dynamics);
- Product data.
Another important factor you ne to consider, and something you really ne to know about your business, is your cost structure. This typically includes the purchase price, as well as all the fix costs associat with each sale. On top of that, you also ne to determine your target gross profit.
Excel is very useful for plotting sales versus price and profit versus price, as well as sales/profit versus time. Remember that the values are not constant.
They change over time as the market evolves and internal and external factors change. For example, for some charts to provide useful information, you may ne to have multiple prices for each SKU in your data set. This could be a standard price and a discount price that you can switch between.
Once you have collect enough data, you can calculate simple statistical key metrics that will allow you to analyze the data:
- Average sales broken down by item per unit of time, e.g. 1 unit sold on average every 10 days => 3 units sold per month on average;
- The difference in sales by SKU, for example if 1 unit is sold every 10 days, this means you have 9 days out of every 10 days where there are no sales.
Using this data and key figures, you can model your sales and profits using:
- Normal distribution (be careful, this may not be the optimal choice for pricing models as it can be negative);
- Gamma distribution (this method is often quite suitable for pricing models);
- Poisson distribution (the method is well suit for pricing models).
Read also our guide to data analysis – How to analyze data: a basic guide .
Why Using Averages Is Not the Best Way to Model Sales and Profits
The mean is a simple statistical figure taken from a list of numbers to represent them. It can be calculat in different ways depending on its use. The mean is primarily us to describe statistical populations that follow a normal distribution (a bell-shap curve), such as population growth.
The arithmetic mean (AM) is calculat as the sum of all the values divid by the number of values in any data set. There are also other types of means such as the geometric mean (GM) and the harmonic mean (HM) which have the mathematical properties of AM ≥ GM ≥ HM in any data set. From now on in this article, we will discuss the use of the arithmetic mean (AM) and simply refer to it as the mean. If a population data set follows a bell curve. The AM has the property of being equal to (the most common data. Point in the data set) and the mian (the 50th percentile in the data set).
CA should never be consider as the only statistical figure when making any decisions because of skewness. In a data set where the vast majority of values are small. Enough large numbers can skew the data set and shift the. Value to the right. If you assume at this point that you are dealing with an Unskew. Normal distribution, your decisions. Will be wrong. The same scenario is true. in the opposite situation, where most of the data points are large and a significant number of small numbers will skew the data to the left.
CA also has difficulty estimating sales where sales occur infrequently. Here are some scenarios that illustrate the problem of using CA in sales forecasting:
Scenario A) 1 sale occurring every 10 days means that there are 3 sales in 30 days (month), 30 sales in 300 days, so 27 days a month there are no sales. The average in this situation will be 1/10 sales per day.
Scenario B) 3 sales occurring on the same day in a 30 day period (one month) means that there will be 29 days in one month with no sales (30 sales in 300 days). The average is still 1/10 sales per day.
Scenario C) 30 sales occurring once in a 300-day period means that there are 299 days without sales during that period. The average is still 1/10 of the sales per day. In All Three Scenarios, the. Calculat Average Is. Far from .Reality, and If You. Were to. Use. This Method, It Would. Result in You. Either Not Having Inventory. When You Ne It, or Having. Too Much Inventory. Now That We’ve Look at the. Problems Associat with. Using an Average to Forecast. Sales and. Profits, It’s Time to Get More. Precise and See. What an. Optimal Allocation Model Would Look Like.