Sales activities in many organizations are manual and inefficient. Sales went into the first call or meeting with little knowledge of the client or why the client contacted them in the first place. Their pitches are not relevant so they lose the client’s attention quickly.
Machine learning can help sales in 3 ways:
Improve Lead Quality
One of the most direct and rewarding way sales team can leverage on data is lead optimization. CRM systems provides advanced lead scoring functionalities which rates a prospect on their action (if it is a Saas business, how they interact on the company website and if a manual sales process, how consumers react); background data (e.g. company size; revenue; consumer demographics) and many other factors.
Per Harald Borgen in his article “Boosting Sales With Machine Learning” explain how they use natural language processing to qualify leads. It uses FullContact API to read description of millions of companies from which it got a full list of company information data (FullContact only accept URL as query input so they have to Google to find the right URL first). Good data were taken from the company’s existing client list and used to train the machine to identify patterns of a good lead.
Prepare For First Meeting
First impression is critical. Most of the articles talked about how important it is to understand the client and the client business, and the client’s competitors and peers. This is a rather manual, time-consuming task at the moment, Clients will find you trustworthy and friendly is you can show you know a little bit about your business.
Moreover, information on client search can also be helpful if client reached out to you from online. Say, if the client has navigated to a specific product information page, then s/he may be interested to understand more on that particular product; if the client download a business case or white paper from your website, then the sales can use business cases focussing on those sectors.
Improve Pricing Outcome
Rakow and Sora talked about how to apply Machine Learning in Motor Insurance Pricing. With new data sources, auto insurers are able to relax some assumption as compared to traditional model and can test different models at the same time. It also means reduction in the number of consumer inputs needed to give a quotation which, for the sales team in the organization, means shorter sales cycle and better sales outcome.
One potential use case comes from the auto insurance industry. New innovation such as electric vehicles, self-driving cars have changed the ways we interact with our cars. But the insurance pricing models aligned with these new technologies are not in place. Virginia Tech study shows self-driving cars are safer than human-driven vehicles at all severity levels. So does having a self-driving car give the owners lower insurance fee? Or actually does how often the car is on the road, which country the vehicle is used also play a role in the pricing? All these new data sources help insurances to improve their pricing model.
Machine learning is not there to replace people, but to empower people. It is up to you and your organization to say if you want to embrace it for the upside and do your best to avoid the downside.
But one thing I bet you will for sure find useful, it can be used to detect your drinking level, to be specific, it helps you prevent sending tweets you wish was “un-tweeted” if you had not been so drunk. So you can rest assured the next time you are as drunk as fiddler that your clients will not become ex-client after you wake up the next morning.
Reference:
Lead Score O[ptimization Machine Learninghttps://www.linkedin.com/pulse/lead-score-optimization-machine-learning-matt-barnes-mba
Application of Machine Learning in Motor Insurance Pricing
https://www.actuaries.org.uk/documents/a1-application-machine-learning-motor-insurance-pricing.
Automated vehicle Crash Rate Comparison Using Naturalistic Datahttp://www.vtti.vt.edu/featured/?p=422
Analyzing Competitor Tariffs With Machine Learninghttp://uk.milliman.com/insight/2015/Analysing-competitor-tariffs-with-machine-learning/
Machine learning can help sales in 3 ways:
Improve Lead Quality
One of the most direct and rewarding way sales team can leverage on data is lead optimization. CRM systems provides advanced lead scoring functionalities which rates a prospect on their action (if it is a Saas business, how they interact on the company website and if a manual sales process, how consumers react); background data (e.g. company size; revenue; consumer demographics) and many other factors.
Per Harald Borgen in his article “Boosting Sales With Machine Learning” explain how they use natural language processing to qualify leads. It uses FullContact API to read description of millions of companies from which it got a full list of company information data (FullContact only accept URL as query input so they have to Google to find the right URL first). Good data were taken from the company’s existing client list and used to train the machine to identify patterns of a good lead.
Prepare For First Meeting
First impression is critical. Most of the articles talked about how important it is to understand the client and the client business, and the client’s competitors and peers. This is a rather manual, time-consuming task at the moment, Clients will find you trustworthy and friendly is you can show you know a little bit about your business.
Moreover, information on client search can also be helpful if client reached out to you from online. Say, if the client has navigated to a specific product information page, then s/he may be interested to understand more on that particular product; if the client download a business case or white paper from your website, then the sales can use business cases focussing on those sectors.
Improve Pricing Outcome
Rakow and Sora talked about how to apply Machine Learning in Motor Insurance Pricing. With new data sources, auto insurers are able to relax some assumption as compared to traditional model and can test different models at the same time. It also means reduction in the number of consumer inputs needed to give a quotation which, for the sales team in the organization, means shorter sales cycle and better sales outcome.
One potential use case comes from the auto insurance industry. New innovation such as electric vehicles, self-driving cars have changed the ways we interact with our cars. But the insurance pricing models aligned with these new technologies are not in place. Virginia Tech study shows self-driving cars are safer than human-driven vehicles at all severity levels. So does having a self-driving car give the owners lower insurance fee? Or actually does how often the car is on the road, which country the vehicle is used also play a role in the pricing? All these new data sources help insurances to improve their pricing model.
Machine learning is not there to replace people, but to empower people. It is up to you and your organization to say if you want to embrace it for the upside and do your best to avoid the downside.
But one thing I bet you will for sure find useful, it can be used to detect your drinking level, to be specific, it helps you prevent sending tweets you wish was “un-tweeted” if you had not been so drunk. So you can rest assured the next time you are as drunk as fiddler that your clients will not become ex-client after you wake up the next morning.
Reference:
Lead Score O[ptimization Machine Learninghttps://www.linkedin.com/pulse/lead-score-optimization-machine-learning-matt-barnes-mba
Application of Machine Learning in Motor Insurance Pricing
https://www.actuaries.org.uk/documents/a1-application-machine-learning-motor-insurance-pricing.
Automated vehicle Crash Rate Comparison Using Naturalistic Datahttp://www.vtti.vt.edu/featured/?p=422
Analyzing Competitor Tariffs With Machine Learninghttp://uk.milliman.com/insight/2015/Analysing-competitor-tariffs-with-machine-learning/