Debt collection in the Energy and
Utilities (E&U) industry is the
proverbial double-edged sword.
Utility providers' attempts to
recover bad debt pose
significant challenges in terms
of customer perception,
engagement and loyalty.
To further complicate matters,
very few customers choose to
engage with their providers to
resolve debt-related issues.
In fact, they are more likely to
feel that they are unfairly
pressured to repay their debts by
unsympathetic E&U companies.
This adversely impacts not only
customer experience and loyalty, but companies' financial
performance as well.
Revenue and profits take a
beating as utility providers
routinely write off bad debts
rather than be constrained by
efforts of failed collections
and an exacting regulatory
environment.
Regulators across the globe
are training their sights on
multiple aspects of consumer
collections — including first-party
debt collectors, borrower
communications, collection of
disputed and time-barred debt,
and audit trails.
In spite of using reputed and
regulated debt collection agencies,
debts in E&U companies have been
on the rise. According to a 2017-18
PwC study, there has been a four
percent compound growth in net
debt levels relative to revenues.1
An integrated approach to debt
collection is therefore the need of
the hour. E&U companies must look
to balance the needs for proactive
identification and management of
high-risk customers on the one
hand, and enhanced customer
experience on the other.
Analytics provides just the right
support.
Driving Payment Behaviors Through Predictive Analytics
With regulators severely restricting
E&U companies from turning away
customers defaulting in bill
payments, companies should
devise innovative methods to
eliminate negative fallouts arising
from wrong actions or even
inaction. Identifying and predicting
payment conditions and ensuing
customer behaviors (and developing supportive plans to
keep their accounts current) thus
becomes critical.
Data literally becomes the fuel for
such an exercise. Internal data
from companies' systems and
external data from social media are
rich inputs for risk-based
evaluation of customer accounts. And predictive analytics can add
muscle to optimize bad debt
collections. Through the right
segmentation of customers,
predictive analytics can provide
accurate insights into customers'
intent and ability to pay. For
example, which customers will pay
without reminders? Who will
respond to early intervention versus reactive efforts? Companies
can effectively map their resources
across the different customer
segments to determine suspension
or termination of services, or even
repayment assistance programs.
This is exactly what a U.S. gas and
utility company has done. They
leveraged a predictive modeling
solution toward forecasting and
optimization, integrating data
management, and creating a more
streamlined collections process.
An analytics-driven strategy
enables efficient pre-delinquency management of debt without losing
out on customer engagement. E&U
companies can reach their
customers ahead of payment time
with individualized messages
through the right channels to
influence payment behaviors.
For example, a leading E&U company2 increased its debt
collection by 50 percent using a
propensity-to-pay predictive data
model and decreased operating
costs by 20 percent. Assigning a
propensity-to-pay score to every
customer, the company predicted the likelihood of customers being
able to pay their dues after their
accounts were finalized.
Accordingly, the company designed
focused delinquency management
strategies and prioritized different
customer segments for specific
actions.
The next generation of smart
meters is poised to flood E&U
companies with sharper and better
quality data that will improve
capabilities toward optimizing
collections. E&U companies can
design creative customer payment plans and proactively address
potential concerns and issues with
real-time information on
customers' changing usage
patterns. Complaints around
inflated bills and bad debts can be
quickly resolved, and companies
can positively interact with their
customers.
Predictive analytics is really all
about visualizing, segmenting and
prioritizing customer
conversations. It efficiently
segments customers for
differentiation and prioritization
much before they become a risk.
Additionally, they can measure engagement levels, usage patterns
and the cost-effectiveness of
channels. Credit counseling
services may also be put in place to
further raise levels of consumer
satisfaction.
The ultimate objective is to have a
robust database of customer
expectations, usage and
satisfaction. The good news is that
every E&U company has billing and
payment data to start this journey.
A consumer's billing history, credit,
payment and collection patterns
are good starting points to design
processes and models for efficient
debt collection.
Data science allows E&U
companies to proactively manage
potential high-risk customers
with confidence. On a reactive level,
it allows for customer
segmentation for debt recovery
with customized corrective
actions. Companies can engage
better with their customers
for effective early arrears
management and rehabilitation.
Users' payment behaviors with
other suppliers can be fed back
into segmentation algorithms for
better predictive models and
more accurate personalization
of collection strategies.
Influencing Customer Sentiments with Social Media Analytics
If predictive analytics leverages
customer profiles, past approaches
and trends to create successful
models, social media analytics taps
customers' likes, dislikes and
sentiments to determine the kinds
of communication and
conversations that will be most
effective. Will the consumer be
more responsive to a phone call,
e-mail or a letter? Who will pay
larger amounts? Who will require
payment assistance?
Intelligence from structured and
unstructured social media data can
be analyzed just as effectively for
debt management as it is done to
improve brand awareness or
reputation management. These
include positive, negative and even
neutral conversations and
comments posted on social media
platforms and other blogging sites,
customer care notes and customer
survey responses.
Sophisticated analytic tools
harness the power of social media
data and integrate real-time
insights [with Business Intelligence
(BI) and Customer Relationship
Management (CRM) platforms] to
sift the customer voice from the noise and facilitate proactive
decision-making.
Braintree Electric Light
Department in Massachusetts,
U.S., uses a social engagement
platform to get customer
awareness and sentiment
feedback, real-time weather
events and outage reports,
logistics support and analytics.3
Social business analysts will need
to coordinate with business
decision-makers to create the
right social pulse relevant to
organizational goals. By integrating
social media customer data with
Enterprise Resource Planning
(ERP) and CRM applications,
business managers can get holistic
BI for the right actions, initiatives
and improvements.
Additionally, companies can develop
social media apps that motivate
consumers to use self-service
channels. This promotes two-way
communication and information
flow from the social media page to
the company's page and vice versa,
besides providing valuable data to
improve customer experience in
billing and collections.
Framework for Social
Media Analytics
- Crawl social media
sites and consumer
forums
- Capture relevant
conversations and
posts
- Extract key
information on
products and
services
- Use analytical
models and
sentiment analysis
to determine
business context and
insights
- Consolidate and
monitor for
continuous insights
An Intelligent Path for Value-added Business Outcomes
Having the right data and systems
to access, connect, understand and
manage huge volumes are real
challenges for E&U companies.
This requires the right talent, tools,
accurate insights and the ability to
execute timely actions.
The value chain is complete only
when insights are transformed into
effective business processes for
multi-channel customer
engagement that influences
payment behaviors.
Predictive and social media
analytics are a formidable
combination. Predictive analytics
closely dissects data elements to
extract patterns that predict the
future behaviors of customers on an
individual basis.
Social media analytics provides the
understanding and inputs for
executing messages that will drive
customers toward expected
behaviors. It enables the lessons
from the predictive models to be 'right-channelized' for relevant,
targeted and prioritized
messages for multi-touch
customer communication
and engagement.
Converging the predictive and
social analytics layers empowers
businesses to ask the relevant
questions, look for the right
answers and take decisions. In
doing so, they can strike the
winning balance between revenue
and customer goodwill.
References:
1. https://www.pwc.com/gx/en/working-capital-management-services/assets/working-capital-opportunity-2017-2018.pdf
2. /insights/case-studies/casestudydetail/223/how-a-leading-utility-company-increased-its-debt-collection-by-50-percent-withpredictive-
analytics
3. http://www.tdworld.com/asset-managementservice/utility-uses-social-media-connect-customers