Development of leading indicator metrics in telesales environment yields 79% improvement in sales and
42% reduction in cost per sale
SITUATION: A Fortune 100 financial services company was using an outsourced telesales channel to sell to companies with revenues less than $10 million per year. Although the call center had been in operation for more than a year and a half, and employed almost two hundred skilled sales agents, it had never met its sales goals, and its cost per sale and ROI were just barely acceptable.
The status quo process involved delivering relatively large numbers of suitable targets, and releasing the entire batch to the call center en masse. At any one time, 8-10 distinct campaigns were deployed in the call center, each with a specific message, offer and product details.
The call center managers were charged with managing the multiple campaigns and designing a process to optimally apply their resources to meet sales goals. At the direction of the call center operations managers, a practice developed whereby the latest and greatest intuition trumped all previous latest and greatest intuitions. As a result, the overall program suffered from:
-
Perpetual incomplete results for campaigns that made it difficult to accurately ascertain true performance or performance drivers;
-
Constant context switching by the agents from campaign to campaign, which limited the agent’s ability to climb a learning curve on any specific message or offer;
-
A sense of insecurity in the mind of each agent about whether they were working on the best campaign relative to their individual commission goals;
-
Above industry standard agent attrition, which put stress on recruiting and training processes.
All of the above contributed to a major problem for the client marketing managers – they could not predict the future results of the program. Because they could not accurately forecast, they were having difficulty making an argument for continued funding of the channel as a whole.
Before, campaigns were not finishing in a predictable way because there were too many operational variables.
ASSIGNMENT: Massini Group was tasked to build an operations framework, including metrics and campaign structure, to bring fact-based decision-making, leading indicator metrics and accurate forecasting into the management of program. The client’s stated objectives for this effort were:
-
Accurate reporting of work in progress, driving...
- Greater consistency / predictability, at... Lower cost, resulting in... Increased sales.
KEY STRATEGY: Massini Group began the process by working to accurately characterize what was actually happening within the program. Related to this effort, Massini Group found the reporting supplied by the call center to be incomplete and unusable, because it provided a large number of small scope, disconnected glimpses into the bigger picture, but no single unified view.
Massini Group decided to build a comprehensive view of the program by constructing a reference database containing:
-
Target sites and contacts provided to the call center by the client;
-
Contacts collected via the calling effort by the call center;
-
Outcomes (dispositions) applied to each opportunity by the call center;
-
Record of the dials, by number and duration, applied by the call center to each opportunity over time;
-
Tags identifying the discrete campaign to which each opportunity and call was attached.
Further, Massini Group analyzed the data based on a set of metrics, proven in 100’s of similar programs managed by Massini Group, to yield cause and effect relationships.
These metrics fell into two categories:
-
Operational parameters or measurements of how resources were applied to the effort;
-
Performance metrics specifically associated with developing a view of work in progress necessary to maximize the results of the agent’s efforts.
ENGINEERED PROCESSES: Massini Group first utilized a formalized process analysis, based on Massini Group’s Dialogue StrategySM, to identify a number of performance-robbing process deficiencies, including:
- 36% of agent time was spent deciding who to call as opposed to actually making calls;
-
42% of the targets received little or no attention from the program;
-
60% of agent calling time was spent on targets that never did buy;
-
29% of agent calling time was spent on targets for which the final outcome was never known;
-
35% of targets were disqualified from any further treatment;
-
The program averaged 82% of its monthly sales targets over the time period.
Massini Group then embarked upon an ambitious plan, with its client’s approval, to progressively rebuild the marketing process one element at a time to eliminate the above deficiencies.
The rest of this paper describes the first installment of a series of marketing process improvements, including:
-
Instituting the use of operational parameter calculators designed to align the number of sales agents and number of available targets, so that each agent had just enough work to complete each month based upon best practices metrics. Right-sizing the sales team relative to the number of available targets assured that all targets received the intended treatment and therefore had a reasonable chance of becoming a sale. This effort alone increased the sales results by 79% and reduced cost per sale by 42%.
-
Instituting enforcement of specific operational parameters designed to control the application of resources to targets, such as, maximum attempts per target resulting in a predictable average attempts per target, minimum dials per hour, and minimum hours dialing per day. Through this effort, it was determined that the team was far too large for the number of targets available if a reasonable and uniform treatment was applied to each. As a result, the improvements in total sales and cost per sale were maintained over the full fiscal year by assuring that the resources available matched the number of targets available.
-
Development of statistically representative batches of targets to assure that all agents within the large team received the same opportunity to complete sales, and that all campaigns had the same opportunity to succeed given a limited number and type of targets.
- This principle was employed for a number of reasons, including:
a. the need to eliminate agent insecurity that they were not each getting a fair shake, which in turn reduced attrition and helped maintain concentration on the task at hand;
b. the desire to institute an organizational learning initiative in which specific tactical and operational alternatives could be tested in a way that assured elimination of statistical bias from the process;
c. the goal that specific metrics could be applied to each subset of targets for the purposes of enhancing forecasting accuracy.
The program was broken down into four functional units each month, resulting in 48 opportunities to optimize the tactics each year. Using this framework, the team was able to:
a. isolate specific tactical or operational issues within days instead of months, as previously was the case;
b. fully characterize the expected performance of previously untouched, recycled records due to maximum attempts and recycled due to non-Interest sub-populations;
c. test specific messages, offers and product configurations in a champion challenger process;
d. document specific seasonality effects.
Ultimately, the leading indicator metrics derived from this process allowed the team to forecast sales in advance within ±5%. Additionally, 13 other intermediate leading indicator metrics were developed to assign specific causes and effects to components of the program, allowing the marketing managers to actively manage many aspects of cost and performance.
RESULTS: The newly engineered marketing process increased sales by 79% and reduced cost per sale by 42%. Additionally, the improvements in total sales and cost per sale were maintained over the full fiscal year by assuring that the resources available matched the number of targets available. Construction of a set of leading indicator metrics, combined with a uniform method for coding and collating historical results, allowed the client company to predict performance, based upon the nature of the targets, to within 5% on a routine basis. The combination of sales results above forecast, sales cost below forecast, and predictable results, assured that the managers of the program would be able to get funding into the future and continue to optimize the process.
After, improved data quality and optimized telemarketing processes produced more predictable and consistent sales from each campaign.
Find out how to bring accountability to every marketing campaign, with metrics that impact the future as well as report on the past in this Using Measurement to Optimize Your Marketing Investment white paper.
