A process manager who is judged only on service level will hit service level. They may also burn out the team, overspend on overtime and leave no one ready to step up as team leader. The balanced scorecard exists to stop one number from crowding out everything else a manager is responsible for.
Where the idea comes from
Robert Kaplan and David Norton introduced the balanced scorecard in the Harvard Business Review in 1992. Their argument was that financial results are lagging indicators: they report what already happened. To manage a business, you also need measures of what drives future results. They proposed four perspectives:
- Financial: how do we look to those who fund us?
- Customer: how do customers see us?
- Internal process: what must we excel at operationally?
- Learning and growth: can we keep improving and building capability?
Kaplan and Norton designed it for organisational strategy, but the same structure works well at the level of an account or process manager in a BPO, where the tension between client delivery, margin and people is constant.
Translating the four views to a BPO account
In an outsourcing setting, “customer” usually means two parties: the client who signs the contract and the end customers who call. Many scorecards treat client contractual metrics as the customer perspective and put end-customer measures like CSAT alongside them. A typical mapping:
- Financial: gross margin of the account, revenue against plan, billing leakage, overtime cost.
- Customer and client: contractual SLAs such as service level and client-audited quality, CSAT, client escalations.
- Internal process: schedule adherence, forecast accuracy, first contact resolution, internal quality, transaction accuracy for back-office work.
- People, learning and growth: attrition, absenteeism, training hours, engagement score, number of team leader successors ready.
Keeping the total to around eight to twelve measures is wise. A scorecard with thirty metrics is not balanced, it is unreadable, and managers will quietly focus on whichever three their boss mentions most.
A worked example
Here is an illustrative monthly scorecard for one process manager. Achievement is calculated as actual ÷ target for metrics where higher is better and target ÷ actual where lower is better (attrition), capped at 110% so one exceptional metric cannot mask others.
| Metric | Weight | Target | Actual | Achievement | Weighted |
|---|---|---|---|---|---|
| Gross margin (%) | 20 | 30 | 28 | 93.3% | 18.67 |
| Revenue vs plan (%) | 10 | 100 | 104 | 104.0% | 10.40 |
| Service level (%) | 15 | 80 | 76 | 95.0% | 14.25 |
| Client-audited quality (%) | 15 | 92 | 94 | 102.2% | 15.33 |
| CSAT (%) | 10 | 85 | 83 | 97.6% | 9.76 |
| Monthly attrition (%) | 15 | 4 | 5 | 80.0% | 12.00 |
| Training hours per agent | 5 | 8 | 9 | 110% (capped) | 5.50 |
| Team leader successors ready | 10 | 3 | 1 | 33.3% | 3.33 |
| Total | 100 | 89.24 |
The overall 89.24 hides the most interesting fact on the page: the manager has prepared one successor out of three. That line contributes only a few points to the total, so a manager can ignore it and still score well. This is the central weakness of weighted composite scores.
Design choices that matter more than the weights
Gates. Some organisations add gates: if client-audited quality falls below a floor, or a compliance breach occurs, the overall rating cannot exceed a set level regardless of other metrics. Gates stop the “great margin, terrible compliance” outcome from looking acceptable.
Caps and floors on achievement. Without a cap, a manager who doubles a low-weight target can offset a missed critical one. Without a floor, a single disastrous metric can produce a negative or meaningless contribution. Choose both deliberately.
Direction and definition. For “lower is better” metrics, decide the formula in advance. Attrition of 5% against a target of 4% gives 80% using target ÷ actual. Other methods, like a fixed deduction per point over target, give different results. Neither is wrong; switching between them mid-year is.
Controllability. If the client cut volumes and revenue fell, should the manager be penalised? Many scorecards measure revenue against a volume-adjusted plan, or move uncontrollable items out of the scored section and into a commentary section.
Leading and lagging, and why the bottom half matters
The real value of the four perspectives is the cause-and-effect chain Kaplan and Norton emphasised. In a BPO, the chain often runs like this: training and coaching (learning) improve first contact resolution and adherence (process), which improve service level, quality and CSAT (customer), which protect revenue and reduce penalties and overtime (financial). People measures are the earliest warnings. An account with rising attrition and falling engagement may meet every SLA this quarter while building the conditions for missing them next quarter.
That is why a manager who consistently hits service level with very high team attrition is not, on a balanced view, performing well. They may be borrowing from the future, using overtime, pressure and goodwill that will run out. A scorecard that makes this visible is doing its job; one that rolls it into a single green number is not.
Using it in a review
The most productive monthly reviews spend little time on the total and most of it on the red items and the trends. Three months of each metric side by side reveal more than one month’s composite. A short commentary from the manager for each red item, covering cause, action and expected recovery date, turns the scorecard from a report card into a management tool.
Written by the RP Strativa Team. Targets, weights and figures are illustrative.
Leave a Reply