Mapping benefits back to business changes and enablers

Chapter 6 found where the gap comes from: the ironers and dispatch for delivery and much of the cost, the wash room for water and energy. Knowing the causes is not yet a plan. Between a cause and a closed gap sit two kinds of change. Some are changes in how the business works: when vans are loaded, in what order linen is ironed, when customers hand over soiled linen. Others are the equipment, systems and contracts that make those working changes possible. Benefits come from the first kind. The second kind only makes them possible.

This chapter maps the targets back to both kinds of change with a benefits dependency network. It gives every benefit and every change an owner, puts the costs that the changes impose on people into the same picture as the gains, and shows that the change which finally carries delivery over the 98% line costs no capital at all.

What a benefits dependency network shows

The benefits dependency network comes from Ward and Daniel’s work on benefits management, developed from studies of information systems investments that delivered the technology but not the gains. Peppard, Ward and Daniel (2007) summarised its central argument: technology on its own delivers no benefit; benefits arise when the organisation changes how it works, and the technology is valuable only as far as it enables those changes. The network makes that explicit in four linked columns.

• Investment objectives: what the organisation must achieve, here the four targets derived in Chapter 5.
• Benefits: the measurable improvements that would achieve them, each with a measure, an owner and a size.
• Business changes: the permanent changes to working practices, roles and relationships that produce the benefits.
• Enabling changes: the one-off changes, such as equipment, systems, training and agreements, that make the business changes possible.

The network is built from right to left. It starts with the objectives and asks, for each, which benefits would achieve it, which business changes would produce those benefits, and what would have to be put in place for those changes to happen. Building it the other way, from the equipment on a supplier’s quote towards the objectives, is how the laundry arrived at a request for a new plant.

Australian public agencies are expected to work this way for some investments: the Digital Transformation Agency’s Benefits Management Policy applies, unless an exception is granted, where a non-corporate Commonwealth entity brings forward an investment with digital or ICT costs. A family business has no such obligation, but the logic is the same.

Building the laundry’s network from the targets

This is an illustration. All amounts are Australian dollars excluding GST. The laundry’s investment objectives are its four targets: at most 93 failed deliveries in the 26-week evidence window, cost at or below A$1.66 per clean kilogram, water at or below 10.0 L/kg and energy at or below 5.0 MJ/kg.

Working back from the delivery target, the operations manager listed the benefits needed in the language of Chapter 6’s causes: fewer failures from peak-day backlog, fewer from ironer breakdowns, fewer from dispatch, and fewer from late soiled linen. For each, the question “what would have to work differently?” produced the business changes.

• Production planned to delivery cut-offs. The ironers work through hospital flatwork for the 06:00 cut-off first and hotel flatwork for 10:00 second, with a written plan for peak days, instead of ironing in the order the linen leaves the dryers.
• A 03:30 loading shift. Hospital vans loaded by 04:45 and hotel vans by 08:00, from a fixed dock plan, instead of loading as linen becomes ready.
• A collection-time agreement with customers. Soiled linen bagged and caged at fixed times, so it reaches the wash room early enough to be washed, dried and ironed for its delivery slot.
• A preventive maintenance routine on the ironer lines, planned around the production plan rather than done after failures.
• Revised ironer crewing, with fewer people feeding and folding by hand.

The cost, water and energy objectives added a rinse-reuse procedure with checks on rewash rates, and a change in who monitors gas and water use each month. Then came the enabling changes: what has to be bought, installed or agreed for those business changes to happen.

Figure 1Benefits dependency network for the delivery and cost objectives, read from the objective down

• Objective: keep the hospital contract at A$1.76/kgat most 93 failures in the evidence window; cost at most A$1.66/kg
  • Benefit: fewer peak-day backlog and breakdown failures157 and 74 failures today
    • Business change: production planned to delivery cut-offsowner: operations manager
      • Enabler: new ironer line with feeder and folder, replacing line 1 (I)A$1,250,000
    • Business change: preventive maintenance routine on the ironersowner: maintenance leader
      • Enabler: maintenance contract for the new lineA$40,000 a year
  • Benefit: fewer dispatch failures111 failures today
    • Business change: 03:30 loading shift and fixed dock planowner: dispatch supervisor
      • Enabler: route software, extra van, early shift (S)A$140,000
  • Benefit: fewer failures from late soiled linen33 failures today
    • Business change: fixed collection times and caged linenowner: the owners, with each customer
      • Enabler: collection-time agreement (C)no capital; A$12,000 a year
  • Benefit: lower labour cost per kilogram53.5% of cost is labour
    • Business change: revised ironer and folding crewsowner: operations manager
      • Enabler: new line's automatic feeder and folder (I) and a towel folder (F)A$160,000 for F

The water and energy objectives produce a second, smaller network. Its benefits are lower water use and lower gas use per kilogram; its business changes are the rinse-reuse procedure, a monthly review of water and gas per kilogram by the finance manager, and a rule that rewash rates are checked whenever rinse settings change; its enabling changes are rinse-water reuse (W), waste-water heat recovery (H) and three water-efficient washer-extractors replacing the three oldest machines (X).

The enabling changes, listed

Together the two networks name seven enabling changes. They make up the outcome-derived option, Option B, which Chapter 8 scores against the alternatives and Chapter 9 prices item by item.

Figure 2Enabling changes in the outcome-derived option

	Capex	Business change it enables	Objective it serves
I New ironer line with feeder and folder, replacing line 1	A$1,250,000	production to cut-offs; revised crewing	delivery, cost, energy
F Automatic towel folder	A$160,000	revised folding crew	cost
S Route software, extra van, 03:30 loading shift	A$140,000	loading shift and dock plan	delivery
W Rinse-water reuse	A$320,000	rinse-reuse procedure	water
H Waste-water heat recovery	A$210,000	monthly gas review	energy
X Three water-efficient washer-extractors	A$360,000	rinse-reuse procedure	water, energy
C Collection-time agreement with customers	A$0	fixed collection times	delivery

The list has a property the original request did not have: no item is there for its own sake. If a business change were dropped, the enabler that serves only that change would go with it.

Sizing the benefits in the target’s own units

Each benefit on the network needs a size in the unit of the objective it serves, or it cannot be compared with its cost. The finance manager sized the cost benefits from the cost model behind the A$1.711 baseline, one element at a time.

• Labour falls from 42 to 35 full-time equivalents, a saving of A$476,000 at A$68,000 each, less A$45,000 of early-shift penalties: A$431,000 a year.
• Gas falls from 5.1 to about 3.96 MJ/kg, about A$99,900 a year at A$28/GJ.
• Water and trade waste fall from 14.0 to 9.60 L/kg, about A$70,000 a year at A$5.10/kL.
• Maintenance falls by A$50,000 a year: A$90,000 no longer spent on line 1, less A$40,000 for the new line’s maintenance contract.

Against those gains sit the costs the changes add: the capital charge on A$2,440,000 of enablers, A$323,776 a year at the book’s 8% convention; A$40,000 a year more for the extra van and the route software; the agreement’s A$12,000; and A$8,424 of electricity. Netting these against the gains is the work of Chapter 8, which scores the whole option. The point here is that every benefit on the network is a number with an owner, so that a claim made at approval can be checked later against the same measure.

An owner for every benefit and every change

A network without owners is a diagram. Ward and Daniel’s approach gives each benefit an owner, the person who will be held to the measure and who has the authority to make the business changes happen, and each business change an owner, who is responsible for making it stick. In a family business with five managers the roles are obvious once someone asks, but they still need to be written down, together with the evidence each owner will use.

Figure 3Owners and evidence for the laundry's benefits and changes

• Do: Delivery benefit: operations managersigned dockets, failures by cause, weekly
• Do: Production planned to cut-offs: operations managerflatwork carried over at end of shift, daily
• Do: Loading shift and dock plan: dispatch supervisorvan loading times against 04:45 and 08:00, daily
• Do: Preventive maintenance routine: maintenance leaderline availability from the downtime log, weekly
• Do: Collection-time agreement: the owners, with each customercollection times logged at the dock, weekly
• Do: Cost benefit: finance managercost per clean kg, fully loaded, monthly
• Do: Water and energy benefits: finance managermeter and invoice readings per clean kg, monthly
• Ask: Dis-benefits: who owns the costs the changes imposeearly-shift penalties, extra driver, roles that go

Who decides, and how fast, matters as much as who owns. Shared goals, co-location and decision rights in Project Teams and Contracts by the Numbers shows how a shared scorecard and delegated decision limits keep a team working towards one set of measures, which is what the laundry’s managers need between approval and the evidence window.

Dis-benefits belong on the network

A dis-benefit is a consequence of the change that someone sees as a loss. Leaving dis-benefits off the network makes the option look better than it is, and leaves no one responsible for managing them. The laundry’s changes impose these.

• Early-shift penalties. The 03:30 loading shift attracts penalty rates under the applicable award: about A$45,000 a year.
• An extra driver. The second hospital run needs one more driver: A$68,000 a year at the laundry’s loaded labour cost.
• The customer agreement’s running cost. Cages and the collection schedule cost A$12,000 a year.
• Electricity. The new ironer line’s drives raise electricity use from 0.25 to 0.26 kWh/kg, about A$8,424 a year.
• Roles that go. The new line’s automatic feeder and folder replace six people feeding and folding by hand, three a shift, and the towel folder one more; the peak-day overtime of about one full-time equivalent also disappears. With the extra driver, the laundry goes from 42 to 35 full-time equivalents.

The last item is not a cost line, and the network does not decide how it should be handled. It does make sure the owners see it before approval, decide how it will be managed with the people affected, and give it an owner. Benefits move after approval describes why dis-benefits that are not owned tend to be discovered, and argued about, after the money is spent.

The change that costs no capital

Each benefit in the network has an estimated size. The operations manager estimated, for each cause in Chapter 6’s Pareto, what share of its failures the relevant changes would remove: 95% of the peak-day backlog, because the new line takes effective capacity above the peak-day need; 85% of the breakdown failures, because line 1 is retired and the new line is maintained to a routine; 80% of dispatch failures, because most were coded to late loading; 20% of the wash-capacity failures, through better scheduling once the ironers stop blocking the dryers; and 60% of the failures from late soiled linen, through the collection-time agreement. These are estimates, and Chapter 10 tests them against what actually happened.

Without the customer agreement, the other six enablers leave 103.8 failures per 26 weeks, an OTIF of 97.8% (97.78%). That is 10.8 failures more than the 93 the tender allows. The equipment alone does not reach the target.

The agreement removes 60% of the 33 late-linen failures, 19.8 failures, and leaves 84.0, an OTIF of 98.2% (98.21%), 9.0 failures inside the limit. Its cost is A$12,000 a year and no capital: about A$303 for each failure avoided, counting both half-years.

Figure 4Failures left per 26 weeks, with and without the customer agreement

	Today	Six enablers, no agreement	All seven enablers
Ironer backlog on peak days	157	7.9	7.9
Ironer breakdowns	74	11.1	11.1
Dispatch, loading and routes	111	22.2	22.2
Wash capacity	37	29.6	29.6
Soiled linen received late	33	33.0	13.2
Total failures	412	103.8	84.0
OTIF	91.2%	97.8%	98.2%

Two lessons follow. The first is that the network found a change the equipment-led request could never have found, because no supplier sells it. The second is that 98.2% is not comfortable. If the changes removed 9.0 fewer failures than estimated, less than one week’s worth at today’s rate of about 15.8 a week, the laundry would be on the limit. Chapter 10 shows how close the actual result came.

The error: the network that left out the dock crew

This is an illustration. The first draft of the network, prepared by the operations manager with the finance manager, listed the enabling changes and their savings but recorded the loading shift simply as “move loading to 03:30”. Costed from that draft, Option B came to A$1.589/kg, comfortably under the A$1.66 ceiling.

The omission surfaced at the review of the network’s owners. The dispatch supervisor, asked what would actually change for the dock crew, said that the 03:30 start would attract early-shift penalties for everyone on it, and that two hospital runs leaving before 05:00 needed another driver, because one driver could not cover both. Neither cost was in the draft: A$45,000 a year in penalties and A$68,000 for the driver, A$113,000 together.

With both included, Option B’s annual cost is A$5,071,051 and its cost per kilogram A$1.625. It still passes the A$1.66 ceiling, but the headroom is A$108,149 a year, about A$0.035/kg. The omitted A$113,000 was larger than the headroom that remains once it is counted. Had a further cost of that size emerged after approval, the option would have failed its own cost target, and the first draft would have given no warning.

The correction was procedural as well as numerical. Every business change in the network now carries a line for “what changes for the people who do this work”, completed by the owner of the change with the people concerned, and every dis-benefit has a cost and an owner before the network is used to score options.

What the network hands to the next chapter

The network gives Chapter 8 three things. It gives a longlist of changes from which options can be assembled, instead of a single request. It gives each option’s effect on each target, through the benefit estimates. And it shows which changes are cheap and quick, such as the customer agreement and the dispatch changes, and which are expensive and slow, which matters when the evidence window opens in month 6.

Exercise: the price of a dispatch failure avoided

The dispatch changes (item S, with the early shift, the extra driver, the van and the software) cost a net A$171,577 a year, including their capital charge. They remove 80% of the 111 dispatch failures per 26 weeks. What do they cost per failure avoided? How does the answer change if the 7.4 wash-capacity failures per 26 weeks that the new scheduling also removes are counted?

Worked answer. Dispatch failures removed are 111 × 0.80 = 88.8 per 26 weeks, or 88.8 × 2 = 177.6 a year. The cost per failure avoided is 171,577 ÷ 177.6 = about A$966. Counting the wash-scheduling gain, the changes remove 88.8 + 7.4 = 96.2 failures per 26 weeks, 192.4 a year, and the cost falls to 171,577 ÷ 192.4 = about A$892 each. Both are far more than the A$303 a failure of the customer agreement, which is why the agreement was worth pursuing even though it removes fewer failures. Neither number is the real test, though: without the dispatch changes, 111 failures remain, more than the 93 allowed on their own, so they are needed whatever they cost per failure.