Service mandate and decision rights · Lesson 8

Tracing benefits and who bears change costs

16 min read

Lesson objectives

  • Tracing benefits and who bears change costs
Reading tools

The service manager announces that a new form saved two hundred hours. At the same meeting, the contact center requests another employee because correction calls increased. Both figures may be accurate while describing different ends of one journey. Before celebrating benefit, ask where work disappeared and where it reappeared. Benefits tracking connects a change to something valuable for users, then examines the resources and consequences involved. Digital investment planning considers resources and expected results when comparing implementation options. Evaluation asks whether change occurred, why, and for whom. This lesson develops a testable benefit register distinguishing initial estimates, observed benefits, and promises whose realization remains in the future.

Describe one benefit in a testable sentence. “Improve patient experience” is broad; “reduce the contacts needed to change a non-urgent appointment” identifies a task and outcome. Add the group, period, completion channel, and comparison condition. You may need both user time and employee time, because improvement for one can increase burden for the other. Assign benefit ownership to someone capable of tracking the outcome, rather than simply the person signing a funding request. If the team controls intake only, state its impact within that boundary and obtain evidence from the receiving team. This supports a shared account of benefit rather than a contest between departments over the largest apparent figure.

Map benefit from the resource to its actual use. A team acquires a tool, trains employees, and changes one request-handling step. It expects fewer repeated tasks, releasing time for cases that previously waited. Each connection contains an assumption. The tool may be ready while staff return to a manual sheet because its data arrive late. Repetition may decline while saved minutes remain too fragmented for work requiring sustained attention. Test separate links: did behavior change, did burden fall, and was released capacity used? Then identify the outcome leadership intends. This locates a weak mechanism instead of labeling the whole project a success or failure from one figure.

Check measurement units before calculating. If the earlier mean is per completed request and the later mean per started request, the denominator changed. A new report might also exclude complex requests transferred to specialists. Trace them and apply a stable rule. Sample records with visible beginnings, endings, and stages; distinguish active handling from elapsed waiting. A user waiting two days does not imply two days of employee work. Reducing wait may also matter more to users than saving staff minutes. Present both measures with collection methods and limitations. Let another reader reproduce a few records without needing to guess what “time” means in the report.

Look for setup and relearning costs. A form may save work after two months while the first month requires staff training, old-data correction, and user communication. Record those activities separately with timing and ownership. Distinguish startup work from recurring monthly operation. Correction repeated every month belongs in operating demand rather than a completed setup line. Also consider reversal work if the change fails: reentering cases, contacting users, or restoring a message format. This makes comparison between a small adaptable trial and a less reversible broad change more transparent. Figures remain estimates until records or tests provide support appropriate to the decision's scale.

Choose early indicators revealing a weak benefit mechanism. If the hypothesis is that a form reduces correction, monitor returned requests and return reasons. If released hours should reduce backlog, observe completion and age of older work while accounting for new demand. Connecting measures to actions clarifies the response when they change. More returned requests may require revising a form question; unchanged backlog may require reviewing hour allocation. Specify before the test what would weaken the hypothesis. A claim retained regardless of the result is difficult to evaluate. A useful benefit owner welcomes early contrary evidence because it can prevent a defective design being extended to more users.

When comparing periods, record changes that could explain the result. Demand may have fallen after a busy season, staffing increased, or completed-request definitions changed. Such explanations do not erase observed improvement; they limit attribution to the project. Place intended and concurrent changes on a timeline. A suitable comparison site or longer follow-up may help, provided differing conditions are examined. Where that is not feasible, describe a coincident change whose explanation remains unverified. Avoid precise promises based on a short sample excluding holidays, absence, or demand variation. A bounded trial can still be chosen under uncertainty when its rationale and the conditions protecting users during learning are explicit.

Prepare the benefit register for a skeptical decision holder. Each row contains one claim, measure, denominator, period, source, owner, and evidence status. Include added work potentially offsetting savings and comparison limitations. State which decision that row could change. A claim affecting no decision may simply decorate a presentation. Ask a colleague to recalculate a row from its source records. Differences may arise from rounding, duplication, or undisclosed exclusion; reconcile the cause before sharing the number. Preserve the earlier version when correcting it and explain the effect. An honest correction strengthens trust more than defending a claim another reader cannot reproduce.

Ask how time reduction affects work quality. Faster handling can come from removing duplication or skipping an important check. Distinguish them using a sample with easy and difficult tasks. Review correct closure, user understanding, and receipt by the proper owner within service scope. When comparing time, state whether quality was comparable and how that was assessed. Better quality may justify longer handling rather than constituting failure. This teaches that value connects a desired outcome, resources, and consequences rather than rewarding the shortest duration. Individual-care implications require qualified review, while the operational committee can examine process completion and accuracy of service promises.

Explain the result to whoever will use released capacity. Thirty available hours may still be unusable if the alternative work requires different supervision or authorization. Plan transfer around a named task, preparation, and completion measure. Then examine whether waiting work improved or the hours dispersed across incidental activities. This closes the last link in benefit realization and connects investment to the operation the institution funds. Tomorrow, choose one promised time saving and identify the task that actually disappeared. Trace its effects to the receiving team and user. Record what current evidence establishes and what needs testing. The result will be a defensible benefit or an honest hypothesis worth examining.

Lesson summary

Tracing benefits and who bears change costs.