| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 3.38 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
No ano de 2019, a empresa mudou a sua estratégia de obtenção de Leads, sendo estas qualquer tipo de contacto fornecido à empresa por parte de outra empresa ou indivíduo. Esta optou por adquirir um número inferior de Leads, mas com um valor de ticket médio, ou receita média por venda, mais elevado, indo à procura de oportunidades com “maior qualidade”. Em consequência desta alteração surgiu a questão de quais as variáveis que mais influenciavam o resultado da empresa e se o Ticket Médio seria de facto uma dessas variáveis. Assim sendo, foi estudada a relação entre as várias variáveis com o intuito de obter o peso de cada uma na Receita. Para obter tal relação foi realizada uma análise preditiva, onde foi aplicado o método de Regressão Linear Múltipla. Para além disso foi também efetuada uma análise descritiva, através da criação de dashboards no software PowerBI, onde foram explorados e analisados diversos KPIs. Os dados utilizados nas análises foram extraídos do software de gestão Bitrix24 e são relativos ao Customer Relationship Management (CRM) da empresa ERP24. Na análise preditiva foram utilizadas como variáveis de estudo o “Ticket médio”, que corresponde à média de receita de cada venda, as “Deals Perdidas”, referentes ao número de vendas falhadas, a Taxa de Conversão Leads em Deals, doravante “Taxa de Conversão L-D”, relativa à taxa de conversão em vendas, de contactos deixados por empresas ou indivíduos à empresa, e o Tempo de Conversão de Novas Deals em Deals Ganhas, doravante “Tempo de Conversão ND-DG”, que representa o tempo que a empresa demora desde que um potencial negócio é introduzido no CRM até que este se concretize em venda. A análise resultou num modelo constituído pelo “Ticket Médio” e o “Tempo de Conversão ND-DG”, que conseguiram explicar 55.5 % da variação das Receitas. Concluiu-se que a Receita irá variar cerca de 3.5 por cada unidade adicional do “Ticket Médio” e 7488.16 por cada unidade adicional do “Tempo de Conversão ND-DG”. Em conclusão e após análise do modelo criado podemos afirmar que a variável “Ticket Médio”, assim como o “Tempo de conversão ND-DG”, tem um impacto considerável na receita da empresa, tendo assim influenciado positivamente os resultados da empresa à alteração de estratégia adotada. Os resultados levam-nos à conclusão que a empresa deve continuar a apostar numa estratégia de aumento do ticket médio e em negócios que demoram mais tempo a converter, pois são aqueles que geram mais receita, o que acaba por ser 8 contraintuitivo, pois a empresa poderia estar mais preocupada com quick-wins, ou seja, negócios de conversão rápida.
In 2019, the company changed its strategy for obtaining Leads, which are any type of contact provided to the company by another company or individual. It chose to acquire a lower number of Leads, but with a higher average ticket value, or average revenue per sale, going in search of opportunities with "higher quality". As a result of this change, the question arose as to which variables most influenced the company's results and if the average ticket was in fact one of those variables. Therefore, the relationship between several variables was studied in order to obtain the weight of each one in the revenue. To obtain this relationship a predictive analysis was performed, where the Multiple Linear Regression method was applied. In addition, a descriptive analysis was also performed, through the creation of dashboards in PowerBI software, where several KPIs were explored and analyzed. The data used in the analyses were extracted from the Bitrix24 management software and are relative to the Customer Relationship Management (CRM) of the company ERP24. In the predictive analysis were used as study variables the "Average Ticket", which corresponds to the average revenue of each sale, the "Lost Deals", referring to the number of failed sales, the Leads to Deals Conversion Rate, henceforth "L-D Conversion Rate", concerning the conversion rate into sales, of contacts left by companies or individuals to the company, and the Conversion Time from New Deals to Deals Won, hereinafter "ND-DG Conversion Time", which represents the time it takes from the time a potential deal is entered into the CRM until it becomes a sale. The analysis resulted in a model consisting of the "Average Ticket" and the "ND-DG Conversion Time", which were able to explain 55.5 % of the variation in Revenue. It was concluded that Revenue will vary about 3.5 for each additional unit of the Average Ticket and 7488.16 for each additional unit of the ND-DG Conversion Time. In conclusion and after analyzing the model created we can state that the variable "Average Ticket", as well as the "ND-DG Conversion Time", has a considerable impact on the company's revenue, thus having positively influenced the company's results to the change of strategy adopted. The results lead us to the conclusion that the company should continue to invest in a strategy to increase the average ticket and in businesses that take longer to convert, because they are the ones that generate more revenue, which turns out to be counter-intuitive because the company could be more concerned with quick-wins, i.e., quick cash conversions.
In 2019, the company changed its strategy for obtaining Leads, which are any type of contact provided to the company by another company or individual. It chose to acquire a lower number of Leads, but with a higher average ticket value, or average revenue per sale, going in search of opportunities with "higher quality". As a result of this change, the question arose as to which variables most influenced the company's results and if the average ticket was in fact one of those variables. Therefore, the relationship between several variables was studied in order to obtain the weight of each one in the revenue. To obtain this relationship a predictive analysis was performed, where the Multiple Linear Regression method was applied. In addition, a descriptive analysis was also performed, through the creation of dashboards in PowerBI software, where several KPIs were explored and analyzed. The data used in the analyses were extracted from the Bitrix24 management software and are relative to the Customer Relationship Management (CRM) of the company ERP24. In the predictive analysis were used as study variables the "Average Ticket", which corresponds to the average revenue of each sale, the "Lost Deals", referring to the number of failed sales, the Leads to Deals Conversion Rate, henceforth "L-D Conversion Rate", concerning the conversion rate into sales, of contacts left by companies or individuals to the company, and the Conversion Time from New Deals to Deals Won, hereinafter "ND-DG Conversion Time", which represents the time it takes from the time a potential deal is entered into the CRM until it becomes a sale. The analysis resulted in a model consisting of the "Average Ticket" and the "ND-DG Conversion Time", which were able to explain 55.5 % of the variation in Revenue. It was concluded that Revenue will vary about 3.5 for each additional unit of the Average Ticket and 7488.16 for each additional unit of the ND-DG Conversion Time. In conclusion and after analyzing the model created we can state that the variable "Average Ticket", as well as the "ND-DG Conversion Time", has a considerable impact on the company's revenue, thus having positively influenced the company's results to the change of strategy adopted. The results lead us to the conclusion that the company should continue to invest in a strategy to increase the average ticket and in businesses that take longer to convert, because they are the ones that generate more revenue, which turns out to be counter-intuitive because the company could be more concerned with quick-wins, i.e., quick cash conversions.
Description
Keywords
Análise descritiva Análise preditiva Regressão linear múltipla Deals Leads Python PowerBI Descriptive analysis Predictive analysis Multiple linear regression