3 AI trends Industrial OEMs should watch out for in 2023 and beyond

Blog

|

Tejas Dhokane

, 

|

Mar 2023

As the Industrial landscape becomes more complex & competitive, Industrial OEMs are under pressure to find new ways to optimize their operations and get ahead of the game. Artificial Intelligence (AI) has emerged as a powerful tool that can help Industrial OEMs address some of their most important challenges. 

With the advent and rapid mainstream adoption of Chatgpt and similar technologies, AI has gone from a ‘we need it, but not sure how we’d use it’ concept to a practical, purpose-driven tool of choice for machinery manufacturers.

In this blog, we will explore three ways in which AI will rapidly transform industrial OEMs in the years to come:

1. Predictive maintenance:

Predictive maintenance is a foresighted approach to maintenance that uses data analysis and machine learning to identify potential equipment failures before they occur. Predictive maintenance using AI can detect equipment issues before they happen, allowing OEMs to schedule maintenance activities and prevent unplanned downtime.

AI algorithms can collect data from previously recorded service transactions, aftermarket parts & service sales, and other sources to create predictive models that can identify patterns and irregularities that are not easily detected by human operators. These models can predict when a piece of equipment is likely to fail, enabling OEMs to schedule maintenance activities in advance, reducing downtime, and avoiding costly repairs. 

A note on IIoT – While IIoT is still a promising technology, its implementation requires real-world upgrades to existing machines in the field. This is often cumbersome and complex as not every machine is upgradeable or the customer may refuse to upgrade at cost. Without a comprehensive roll-out of IIoT across the entire Installed Base, a narrow view of a few machines with IIoT hardly helps a machinery manufacturer pre-empt service requests and provide the required aftermarket parts & service resources.

2. Aftermarket Sales:

Aftermarket sales are an important revenue stream for industrial OEMs. The aftermarket revenue of an asset can be as high as 5 times the original equipment price over years of operation. However, traditional aftermarket sales methods can be inefficient and may increase costs, as they mostly rely on manual processes and may involve unnecessary inspections or replacements. 

AI is transforming the way aftermarket sales used to work by enabling OEMs to provide personalized and more efficient services to their customers. Example: Artificial Intelligence-powered analytics can be used by Industrials to identify which parts are more likely to fail and when, allowing OEMs to offer predictive maintenance services and sell parts that are more likely to fail, this will also help in inventory management by ensuring that the right part should be available at the right time.

3. Inventory Management:

Effective Inventory Management is critical for the success of Industrial OEMs to enhance the effectiveness of their customer service. However, managing inventory is quite a hassle for the Industrials as it involves balancing the stock needed to be stored in addition to the cost of holding the inventory.

Artificial Intelligence is transforming inventory management by enabling Industrials to optimize their inventory. Artificial Intelligence-powered analytics can analyze historical data on sales and inventory, and also the other factors such as weather or season, political, and market conditions to forecast the demand and optimize the inventory required. This helps Industrials reduce inventory costs and ensures that they have the right part at the right time to cater to their customers.

In conclusion, Artificial Intelligence is transforming the Industrial OEMs landscape by enabling predictive maintenance, optimization of aftermarket sales, and optimizing inventory management. By leveraging Artificial Intelligence, OEMs can increase efficiency, reduce costs, and improve customer satisfaction, giving them a competitive edge in a rapidly evolving market. As AI technology continues to advance, we can expect to see even more transformative applications in the Industrial sector.

About Entytle

Entytle, Inc. provides an Installed Base Automation (IBA) platform that assembles, cleanses, analyzes, and operationalizes Installed Base data so machinery manufacturers can make customer-facing workflows more efficient. Entytle’s IBA platform is deployed across thousands of Industrial OEM users. Other applications on the platform include IB HealthCheck, Customer Loyalty Manager, Data Quality Engine, and Entytle APIs, web and mobile interface amongst others that run on the versatile IBA platform. The cloud-based platform includes purpose-built AI that provides a complete 360 view of the Installed Base, intelligent hunting lists, and the ability to orchestrate automation between various tools, systems, or processes. This enables smarter, faster workflows leading to increases in productivity, capacity, and scalability. Industry leaders such as Johnson Controls, Baker Hughes, Peerless Pump, Dematic, Duravant, GEA, and many more trust Entytle to help drive efficiency and growth using their Installed Base. Learn more about how Entytle can help you win over your Installed Base and drive commercial productivity at www.entytle.com.

Using Data Science to find patterns from Installed Base purchasing behavior

Blog

|

Vrushabhkumar Jain

, 

|

Feb 2023

A primer on how FCRM (Frequency, Consistency, Recency, and Monetary Value) analysis can help Industrial OEMs predict buying behavior.

Customers expect OEMs to understand their needs and expectations and hence they have to come up with the right marketing strategies that work for their customers. Successful organizations do not make product, service, and marketing decisions based on their ideas, but they take into account things such as what products/services customers purchase, how often and regularly they purchase, what makes their customers engage with them, etc. Hence, understanding customer behavior at scale becomes paramount for OEMs in such cases. This is where FCRM (Frequency, Consistency, Recency, Monetary Value) analysis could help.

An analysis of customer FCRM data can help OEMs understand customer purchasing patterns, identify potential problems, and improve customer satisfaction. FCRM analysis help marketers better strategize on their campaigns, fine-tune their efforts and improve their results. Thus, FCRM plays a vital role in learning and gaining insights from the vast amount of customers and their transactional data.

But, What is FCRM Analysis for Industrial OEMs?

FCRM analysis is a 4-dimensional framework that ranks customers based on their FCRM scores. The framework is a behavioral approach to target customers a company can focus on to maintain a healthy business. The analysis help OEMs know who are their best customers and which customers need their attention.

But what exactly is FCRM?

F for Frequency:  How often do the customers engage/purchase?

C for Consistency: How regularly they engage/purchase?

R for Recency:  How recently did they engage/purchase?

M for Monetary Value:  How much did they spend?

Let’s understand these scores with examples. The below table contains the transaction history of few customers between 2015-01-01 and 2017-04-30. These scores can be calculated at different time durations (time ordinal), it can be day, week, month or year. The table below contains time ordinal in weeks and hence contains data for 121 weeks.

Below table contains FCRM scores derived using customers’ above transaction history which are indicative of customers purchasing behavior. Each FCR score contains value between 0 to 1 where 0 is worst and 1 being best and Monetary Value is the total amount spent by customers on orders.

As one can see, Customer 2 is making purchases often, hence, their frequency score is reflective of that. Customer 3 is not making purchases very often, but on a regular basis which reflects its low frequency and high consistency scores and the high score of recency also tells about the recency of its transactions compared to other customers’ last transactions in the same period.

By analyzing the above scores, OEMs can figure out which customers are most valued and can take necessary steps to improve relations with those customers. The below image shows a few examples of types of customers identified using FCRM scores.

Focusing on one type of customer can often lead to neglecting customers of another type. Unsatisfied customers can be less frequent, less valuable, and older than a company’s best customers. Such unsatisfied customers may never complain, but by understanding their behavior, a company can work on improving overall customer satisfaction.

When considering other customer attributes along with FCRM scores, such as the type of product or service they purchase or the industry they belong to, business unit, etc., It’s helpful to divide customers into different groups of similar types. This way, you can more easily conduct exploratory data analysis of each customer group and create customer profiles. These profiles can then be used to identify the needs, requirements, and pain points of certain customer types. From there, you can tailor your marketing efforts to better meet their needs.

Steps to fix Installed Base Data for Installed Base Selling: Data Profiling – Part 2/4

Blog

|

Shrihari Mundada

, 

|

Jan 2023

What gets measured, gets managed – Peter Drucker

Data-driven decisions need data. Data that is clean and accurate. In large organizations, the data quality aspect must also be able to execute at scale. There are data challenges across the volume, which are as well significant. Over the years, data has resided in every part of the organization in all shapes and sizes possible. The availability of the accurate and right amount of data significantly impacts decisions from executives to front-line salespersons.

In the first part, we looked into how we can make a list of data sources and build a relationship diagram between different systems. The second part of the blog will look into how the data can be profiled so we can get a better understanding of various data sources and at what quality level they are.

As an analyst, if you have to get your report/dashboard right for your executives, you must build a bunch of little stories around your data. This is where data profiling comes into the picture. The core advantage of data profiling is that it tells you something about your business attributes without getting much deeper into it, record by record.

But before you start profiling your data, you need to understand the context, because if you want your executives to make the right decisions for your organization, you will have to tell them stories the way they understand. For example, Industrial OEMs usually have the following type of data:

1. Customer Locations
2. Transactional data – Orders
3. Assets
4. Services
5. Service Contracts
6. Warranties
7. Product Catalog

For manufacturers, having an Installed Base of 10K locations and 100K parts and equipment over a period of time, the transactional data would be huge enough to make it impossible to review it record by record. In the real world, all this data would reside in different tools, and spreadsheets and some data may not even exist. So at Entytle, we built a data profiling tool that allows you to merge different files into one and view all data in a single place. As shown in the image below, you can choose 5 different files, merge all of them together and generate a profile of the merged dataset.

Once the profile report is generated, it becomes easier to gauge the quality of the data. You must look for features like Cardinality, Missing Values, Patterns, Outliers, Median values, and Min and Max values in the dataset to gauge its quality. Datasets with tremendous outlier values or missing values would be the obvious ones to drag your data quality down. For example, a dataset having $600Bn of [orderAmount], as shown in the image below, is a huge outlier. If accurate, it could change the fortunes of the manufacturer! One more outlier is 70% of the dataset has a Zero [orderAmount] value. These are just a few examples, but the crux of the matter is, with this data level of quality, it would be impossible for your executives to plan strategies that are in sync with the Installed Base value.

Finally, data cleansing and enrichment include identifying, modifying, or removing any data from your dataset that is incomplete, incorrect, or irrelevant. This is where profiling helps. Moreover, with different types of objects mentioned above (Location, transactions, etc.), the quality metrics differ. There are tons of profiling tools out in the market, but we at Entytle, have built an end-to-end platform that helps you ingest, profile, clean, and unify your data keeping your Installed Base Models at the heart of it. It eventually would help you realize the actual value of your Installed Base. For more information, talk to us at info@entytle.com or contact us.

Read Previous Blog:

Steps to fix Installed Base Data for Installed Base Selling – Part 1/4

AI in Aftermarket for Industrial OEM in 2023

Blog

|

Lalit Bhatt

, 

|

Jan 2023

2022 will be a memorable year in history, not only because of the ongoing pandemic and its aftermath, but also due to the significant advancements in Artificial Intelligence (AI). While AI still has its limitations, the chatGPT technology has demonstrated the potential of large language models and how they can be applied in various aspects of our lives. In 2023, the increasing adoption of AI will bring both opportunities and challenges. Those who are ready to embrace the technology will have a competitive advantage, while those who are slower to adopt may face difficulties.

AI is becoming more mainstream and will have a widespread impact on all aspects of life, both human and non-human. In the past, automation has been driven by the need for speed and processing power, but chatGPT has made large language models more accessible, enabling a wide range of AI applications. This development raises the question of what will happen to companies like Google, which is reportedly treating the situation as a “Code Red” emergency. The impact of chatGPT and other AI technologies will likely be felt most strongly in industries that are yet to fully embrace digital tools and transformation. In fact, the ability to adopt higher paradigms like AI may become a matter of survival for some organizations.

Manufacturers and Industrial OEMs might think that with their scale and reach, they are in a safe position. The problem is the speed and intensity of disruptions that new technologies are bringing. Industrial OEMs do appreciate the need for digital transformation, but it is still not seen as an urgent priority. Most of the industrial OEMs are focussed on their top customers. And many times, they don’t know how to deal with long tails.

• Research by McKinsey has shown that each percentage point by which services grow over product sales correlates to a 50% increase in enterprise value.

We can state a corollary here. When an organization fails to fully monetize its wallet share in the aftermarket, it is allowing others to enter its customer base. This also has implications for future equipment sales. 

Aftermarket is essential for the financial health of an enterprise.  The question, however, comes how we can leverage digital tools and AI for doing aftermarket sales and services.

Let me start with the boring topic of data quality. Most OEMs think that they cannot do anything because their data is a mess, and there is no way to fix it. That is no longer the case. With many AI-driven data quality digital tools, that is easily achievable. With Entytle’s Data Quality Engine, we have seen multiple success stories. We have seen successful data quality projects where the data is cleansed, enriched, deduplicated, and unified. We have seen a 30-50% reduction in data size because of duplication. 100 million + records across assets, parts, and transactions have been unified successfully to build a 360 view of the Installed Base.

Once the data quality is in place, other things fall in place. AI can help with a data-driven approach to aftermarket engagements. It can help with diagnostics as well as data points for future engagements. 

Here are a couple of things where AI can really help out 

Customer segmentation

AI can help with segmentation and clustering customers in multiple ways. This helps in defining strategies around different segments. One of the most effective ways is to segment them around the notion of loyalty. The segmentation can help in identifying drifting customers as well as those who are showing promise for growth. 

Leads and Opportunities

Equipment work on duty cycles. For a given operating condition, the same equipment behaves in a similar way. They have similar needs for consumables, services, and parts replacement. AI algorithms can look into the consumption data patterns into similar cohorts of customers and predict the opportunities. When someone is not buying from you might as well mean that they have gone to a competitor. Time to recapture the wallet share!!

Propensity to buy

Let’s say the AI system delivers two opportunities for $20,000 each. Which one to go after if resources are available for only 1? Propensity to buy can help in that as it will tell the likely probability of the prospect buying it. This helps in prioritizing the efforts. 

Recommended items

The sales team loves to upsell and cross-sell. But which items to bundle together. AI can help with recommendations. It can look into the transaction histories and find out what people buy together. 

Whom to call

Everything is in place. The predictive opportunity is there with the list of recommended items. The propensity to buy shows a high probability, but how to decide whom to call? For top customers, we carry their phone numbers in our pockets but that is not true with a long tail. AI can help in surfacing the contacts, and with the appropriate weightage model, it can create a calling sequence. Based on how the contact was mined, it can even recommend if the contact is appropriate for a service or a part sale.

Summary

Data-driven approach coupled with AI can unlock the value of the installed base in the aftermarket. It helps in building and enriching the connect with the customers which further helps in future equipment sales.

10 things that every Industrial Website should have

Blog

|

Soham Kamthe

, 

|

Dec 2022

As an industrial business, having a strong online presence is crucial for attracting potential customers and showcasing your products and services. Your website is the first thing that your prospective customers see about your company and having a good first impression always plays off in the long run. Now a days Industrial websites are also moving towards becoming modern and adopting ongoing trends. However, there are a few common mistakes that many industrial websites make, which can hinder their success and impact their ability to attract and retain customers.

Here are the top 10 things that every Industrial Website should have: 

1. A Mobile-Friendly Design: In today’s world, more and more people are accessing the internet on their smartphones and tablets. If your industrial website is not mobile-friendly, it can be difficult for these users to navigate and use your site, leading to a poor user experience. This can result in potential customers leaving your site and choosing a competitor’s site instead.
2. A clear and concise explanation of the products or services that the company offers: An example of this would be a section on an industrial website that outlines the different types of machinery or equipment that the company manufactures, along with detailed descriptions of each product’s features and capabilities.
3. A professional and user-friendly design: An example of this would be a website that is visually appealing and easy to navigate, with a clean and organized layout and clear calls to action.
4. Customer testimonials or case studies: An example of this would be a section on an industrial website that features quotes or stories from satisfied customers, describing their experiences working with the company and the results they achieved.
5. Contact information: An example of this would be a page on an industrial website that includes the company’s physical address, phone number, and email address, as well as a form that visitors can use to submit inquiries or requests for more information.
6. Technical specifications and documentation: An example of this would be a page on an industrial website that provides detailed information about the technical capabilities of the company’s products, including technical drawings, data sheets, and user manuals. This type of information is especially important for industrial products, as it helps customers understand the capabilities and limitations of the products and make informed decisions about whether they are a good fit for their needs.
7. A list of accreditations and certifications: An example of this would be a page on an industrial website that showcases any relevant certifications or accreditations that the company holds, such as ISO 9001 certification or UL listing. This can help build trust with potential customers and demonstrate the company’s commitment to quality and safety.
8. A FAQ section: An example of this would be a page on an industrial website that provides answers to common questions that customers may have about the company’s products or services. This can save time for both the company and its customers and help reduce the number of inquiries the company receives.
9. A customer portal: An example of this would be a secure login area on an industrial website that allows customers to access their account information, place orders, view invoices, and more. This can improve the customer experience and make it easier for customers to do business with the company.
10. Contact Us: This section should provide information on how visitors can get in touch with the company, including its physical address, phone number, and email address. It should also include a contact form or other means for visitors to send inquiries or request more information. For example, a company that provides industrial consulting services might have a Contact Us page that includes its office locations, phone numbers, and a form for potential clients to request a consultation.

How to implement an Installed base management solution?

Blog

|

Shraddha Patil

, 

|

Dec 2022

Installed base management is tracking and managing the physical assets, products, or systems a company has installed at a customer site. This can include anything from machinery and equipment to software and other digital products. An installed base management solution is a tool or system that helps companies effectively manage and maintain their installed base, ensuring that customers receive the best possible support and that the company can maximize its installed base’s value.

Here are the 10 steps involved in successfully implementing an installed base management solution

1. Define your business objectives: Clearly define what you want to achieve through installed base management. This will help you determine the scope of your efforts and the resources you need to allocate.
2. Identify your installed base: Create a complete and accurate inventory of your installed base, including information such as product or system types, serial numbers, locations, and installation dates.
3. Set up your installed base management system: Choose a system or tool that will allow you to track and manage your installed base effectively. This could be an Installed Base Platform, CRM, BI tool, spreadsheet, or any other system that meets your needs.
4. Collect and track data: Gather and store data on each asset or product in your installed base, including information such as warranty information, service and maintenance records, and any other relevant data.
5. Set up processes and procedures: Establish processes and procedures for managing your installed base, including how to handle warranty claims, service requests, and other customer support issues.
6. Integrate with other systems: If necessary, integrate your installed base management system with other systems, such as your customer relationship management (CRM) or enterprise resource planning (ERP) system.
7. Train your team: Ensure that your team is trained on how to use the installed base management system and follow established processes and procedures.
8. Communicate with customers: Keep your customers informed about the status of their installed base and any issues that may arise. This can be done through regular communication such as emails, phone calls, or in-person visits.
9. Monitor and track key performance indicators (KPIs): Use reports and dashboards to track key performance indicators (KPIs) and identify trends and patterns in your installed base. This will help you to make data-driven decisions about how to optimize your installed base and improve customer satisfaction.
10. Continuously improve: Regularly review and assess your installed base management processes and procedures, and make adjustments as needed to improve efficiency and effectiveness.

Leading Industrials use an Installed Base Platform for all things Installed Base.
Learn more about an Installed Base Platform or Contact Us.

Tech-Enabled Go-To-Market Innovation for Industrial OEMs in 2023

Blog

|

Vivek Joshi

, 

|

Dec 2022

Industrial companies that modernize their go-to-market strategy with the use of technology, improve their revenue growth as well as customer retention and expansion. Industrial companies face an increasing need to innovate their traditional go-to-market strategy and business model. The whole sector is being disrupted through the rise of digital commerce, competition innovating, and the changing of buying patterns by end users. Yet only some are managing to adapt quickly enough, lagging in digital maturity. 

However, the industrial companies that are moving quickly and decisively to enact digital transformation initiatives will unlock significant value for themselves and their end users. One solution we have seen for tech-enabled go-to-market innovation for manufacturers is enabling your dealers and channel partners with A.I. and predictive selling solutions geared towards driving customer engagement and maximizing the lifetime value of the existing Installed Base. Helping your partners and end users be more successful will drive organic growth and recurring revenues for the organization. 

One Source of Truth for Your Installed Base Data 

To fully capture the revenue uplift from selling, it is critical to combine transactional data across systems from the OEM and your channel partners into one place. This will establish a single source of truth for your sales and service teams. This is often the hardest thing to accomplish, given the multiplicity of data sources, taxonomies, and duplication among the systems. 

This is when you must cleanse, dedupe, and enrich the data to have a clear view of the Installed Base. This is a difficult process; many industrial companies have spent years and millions of dollars with no progress. This also must be a continuing process of data collection, cleansing, and enriching to maintain real-time visibility into the Installed Base. 

“By setting permissions in the data sets, the data is only visible to the dealer or channel partner who owns the data.” 

The ultimate benefit of having one source of truth for your transactional data is the ability to understand your customer at a granular level. With insights into customer buying behaviors in hand, the best manufacturers go for big opportunities. These are comprehensive with the ability to move levers in different areas of the business, including identification of drifting customers for churn reduction, identifying unprofitable service agreements, segmenting the Installed Base for equipment upgrade campaigns, and insights for product management on the life of an asset. 

Entytle Installed Base Platform makes it easy to create a single source of truth for your customer interaction data. We use intelligent permission sets to maintain data integrity and protect visibility for the OEM or their partners who own the data. Our system can be stood up in 8- 12 weeks with cleansed, deduped, and enriched data. We’ve done this for many of the biggest global industrial manufacturers.