Collections Best Practice

Utility bad debt is increasing year-on-year according to market research group
Chartwell in its 2004 report, Credit and Collections in the Utility Industry.
Chartwell estimates that more than $1.7 billion in revenue was written off by
North American utilities in 2003, an average of $8.50 uncollected per customer.
Bad debt ratios for individual utilities can vary widely, with some utilities
having net bad debt as low as 0.15 percent of revenue, while others have seen
their debt rise by millions of dollars per year to well over 2 percent.

In a regulated marketplace the cost of utility bad debt is borne by ratepayers,
but regulators’ tolerance is wearing thin. Many utilities are finding themselves
in the spotlight, having to reduce their bad debt, and it is not uncommon to
see improvements in collections performance listed in the top five corporate

Utility debt is expected to continue its rise amid a sluggish economy and an
environment of rising energy commodity prices, forcing utility collection managers
to revaluate their collections assumptions and processes if they are to make
progress in improving collections performance. Old-school one-size-fits-all
approaches are no longer adequate in today’s challenging economic and regulatory
environment, and collections best-practice concepts are catching on fast.

“Utility collections best practice is about applying collections processes
appropriately to each customer account,” says Bob Cooke, principal of risk strategy
at specialist utility collections consultancy Bass & Company. “It is founded
on a deeper understanding of the credit risk of individual customers and customer
segments and implementing tailored treatment paths that focus resources on customers
most at risk.”

Collections Best Practices

  • Multi-Jurisdiction Compliance – Utilities operating in multiple states or
    countries require wide-ranging collections configurability to meet the varied
    stipulations laid out by each state or country. For example, in some states
    a utility is required to attempt customer contact twice before disconnection.
    In other states, disconnecting certain customer groups during the winter months
    is prohibited.
  • Customer Segmentation – Segmentation of a utility customer base for collections
    activities must extend beyond a traditional credit score or an acknowledgement
    that a customer is either “good” or “bad.” These categories, the basis for
    traditional utility collections practice, are too broad to be effective when
    measured against best-practice concepts. For example, there are customers
    who generally pay on time and occasionally late; others who find it difficult
    to pay and require payment plans; business or government customers who pay
    late but are a very low credit risk; and those who are continually being issued
    termination notices. To classify all four groups as “bad” and treat them equally
    will mean actions are too aggressive for some customers and not aggressive
    enough for others. Best-practice collections will classify customers into
    appropriate segments. In addition, some regulatory jurisdictions require detailed
    customer segmentation. In Pennsylvania, for example, regulations require specific
    collections treatments for “Level 1 and 2” low-income customers during winter
  • Configurable Credit Scoring – Scoring credit risk based on customer attributes,
    historical payment patterns and other factors affecting credit risk as opposed
    to traditional methods that primarily measure payment timeliness. Utilities
    may choose their own business-defined parameters for calculating credit scores
    and may choose to score customers from 1 to 5, 1 to 10, or 1 to 1,000, depending
    upon what best supports their business practice. Credit scores may be adjusted
    based not just on whether customers have historically paid on time, paid late
    or received reminder notices, but also on other factors affecting credit risk
    such as whether the customer provided identification numbers or telephone
    numbers at the time of sign up.
  • Dynamic Action and Reaction – Traditional utility collections operate on
    a cycle basis: Each customer account is checked when its cycle comes around
    each month. Each business day, perhaps 5 percent of the customer base is checked
    for credit and collection status. But the real world is not cycle-based. Circumstances
    can change daily. Customers may make payments or have payments dishonored.
    Customers may provide additional security, their credit ratings may improve
    or worsen, collections policies may be adjusted, or the customer account may
    be reclassified to a different segment. Any of these factors can dictate an
    immediate requirement for collection action. Best-practice collections require
    that collections processes not be bound by billing cycles but by a more dynamic
    action: to act early to collect to minimize credit risk exposure.
  • Tailored Treatment Paths – These are designed to suit the anticipated profile
    of each customer. This might include printed letters, email, text messages,
    automated phone messages, outbound call center calls, account manager escalations,
    disconnection work orders and website messages, when appropriate. These can
    be configured to meet individual customer requirements such as agreed payment
    structures for those struggling to pay, polite reminders for customers who
    normally pay on time but on occasion forget, email reminders for customers
    who have chosen that as their preferred communication channel, and stronger
    warnings and accelerated cycles for delinquent customers who regularly default
    on outstanding debt. Different treatment paths are required to support each
    classified customer segment, within the general requirements for differing
    treatments for types of customer for each regulatory jurisdiction.
  • Outsourcing to Limit Exposure – As the collections process for each customer
    moves further along each treatment path, and as arrears age, the probability
    of recovery decreases – so does the financial value of the receivable (see
    Figure 1). As a customer’s debt ages a utility might look to “factor its debt,”
    meaning sell the receivable to an outside agency or consider whether it should
    simply be written off. Bass & Company research indicates that debt factoring
    can be applied more widely than just for very late-stage delinquencies. For
    example, for higher-risk customers, terminating the account earlier and outsourcing
    it to an agency, depending on the regulatory constraints, can reduce overall
    write-offs as the bad debt exposure is limited by the factoring process. It
    is also important to identify which groups of customers should be outsourced
    to which agencies – some agencies might have more skill and success with business
    customer debt recovery than residential.

  • Manage Cost to Collect – Best-practice collections focus not just on amount
    collected but on optimizing the cost incurred in making those collections.
    Collections costs are incurred in the call center, in letter printing and
    mailing, in working capital costs through delayed receipt of monies, and in
    bad debt written off. For every dollar spent on collections activities, more
    than one dollar should be recovered. By managing collections with an understanding
    of cost-to-collect and options such as debt factoring, collections costs can
    be optimized.
  • Collections Intelligence, Analytics and KPIs – Best-practice collections
    require continuous improvement through an interactive approach. Utility collection
    managers should use targeted collections campaigns and adjustments to collections
    treatment paths, analyzing the impact and regularly refining collection treatment
    paths and customer segmentation based on the results. This continuous improvement
    process requires the ability to monitor collections performance by customer
    segment, and to analyze metrics, key performance indicators (KPIs) and trends,
    and relate them to collections campaign dates. Interactive, what-if analysis
    will allow a utility to model and determine the effectiveness of alternative
    collections policies, monitor such metrics as “bad debt percent to revenue”
    and “days sales outstanding” and gauge their impact on operational costs.
    In addition, advanced data analysis using metrics and trend displays, and
    collection control dashboards that illustrate KPIs such as status of overdue
    receivables and corresponding collections actions, will provide the necessary
    business intelligence to adapt and extend collections activities.


A forward-thinking utility responding to the industrywide best-practice collections
challenge is Xcel Energy, the fourth-largest electricity and gas utility in
the United States. Xcel Energy required parameter-configured collections software
that would allow it to move away from static customer collections profiles that
pigeonhole customers into oversimplified “good” or “bad” categories in order
to deliver targeted credit and collections treatments. Today, Xcel Energy is
able to advance beyond the typical utility collection model of standard reminder
letters, customer calls and disconnection notices generated for all overdue
customers on the same number of days after due date.

Xcel Energy’s service territory covers 10 states, requiring wideranging collections
configurability to meet the varied regulatory stipulations. In New Mexico, for
example, collection cycles are allowed to be far more aggressive than in Colorado,
and it makes little business sense to implement the same collection cycles in
both states. Figure 2 contrasts credit treatment processes in Colorado and New

In Colorado those customers with a “good” credit score receive a reminder letter
33 days after their due date, a notice of disconnection on day 64 and the disconnection
process starts on day 74.

Customers in the “others” category do not receive a reminder letter at all,
only a notice of disconnection on day 33, and the disconnection initiates on
day 41. In New Mexico, by comparison, collections treatments differ for commercial
and residential customer segments. The residential customer has 10 days more
between their invoice date and due date, and the timeframe for disconnection
initiation is much shorter for a commercial versus a residential customer: 10
and 39 days, respectively.

Removing the Systems Barrier

Cooke at Bass & Company believes there is a basic understanding in the utility
industry of the concepts of collections best practices, but almost all utilities
lack the technology to be able to adopt best practices. “Everywhere we have
taken our collections models, technology has been the limiting factor,” he says.
“Without the right systems in place, collections best practice is an unobtainable

Traditionally, collections has been the technology domain of the utility’s
customer information system (CIS). Many utilities struggle with weak and restrictive
collections functionality in an aging CIS. Others may have a relatively new
CIS in place that boasts adequate collections support but is missing key features
to support today’s best-practice collections processes.

Utilities collection managers, dependent upon the CIS platform to manage collections
processes, have been resigned to waiting years, perhaps decades, for a full-scale
CIS replacement opportunity so they can implement improved collection processes.

One might suppose that good, paying utility customers are doomed to subsidize
bad debt as utilities are resigned to absorbing write-offs above regulatory
allowances until the day they can do a full-scale CIS replacement. But this
is no longer so with the advent of new software products that can integrate
with an existing CIS and deliver muchneeded collections capability to support
best-practice approaches.

Utility debt is not going away. There will always be customers who default
on payment. But technology has caught up with the problem, and utilities now
have the option to implement specialist collection software products that enable
best-practice approaches.