I’m a “fix or repair” guy.

I differentiate myself from “throw away and buy new” guys. I have a tendency to drive cars beyond what others might call “their useful life.” I own too many things held together with twisted wire and duct tape. I get unreasonable pleasure from improvising a solution with available materials in my shop rather than making another trip to Home Depot.

As such, I imagine myself to be a good guy to have on a backpacking trip, because if equipment breaks when we’re in the middle of the woods miles from REI or Eastern Mountain Sports, I can always “jury-rig something up.” I’m proud of this skillset, but it has its limits.

In times of change businesses all-too-often default to the quick fix. “Can you patch the software?” “Do a workaround,”  “I know we really need to hire someone, but give it to Mary she’ll figure out how to get it done.” I understand the allure of the improvised solution, the quick fix, but it isn’t always a good idea.

Years ago, I worked for some clients in oil and gas exploration and production in the western US. I used to hear things like “Time to cowboy up,” or “Git’er done!” You don’t hear that so much anymore – too many process safety incidents have claimed lives.

In VUCA be Agile

There has been much written about the constant change of the current business environment. It is described as Volatility, Uncertainty, Complexity, and/or Ambiguity  (VUCA). VUCA was coined by Warren Bennis and Bert Nanus in 1987. The US Army War College picked up the term in the 1990s for use in military strategy.[caption width=”276″ id=”attachment_2332″ align=”alignnone”

In 2014  Nate Bennett and G. James Lemoine in a Harvard Business Review article, recommended analyzing which was the dominant condition and acting accordingly. For example, build in slack and redundancy to cope with volatility or collect data and experiment to deal with uncertainty and ambiguity.

Unfortunately the language of VUCA has found greater adoption with many leaders than the solutions that Bennett and Lemoine recommended.

Another sign of the change affecting leaders is the methodology Agile. Agile was originally an idea to speed up the software development cycle. Software development was a serial process  -concept – design- develop- test. This was sometimes called the waterfall process as it flowed down and you had to complete one phase before starting the next, which was time consuming..

Agile was a series of quick design-develop- test loops (sprints) by multidisciplinary teams who met daily (scrums) to get to rapid prototyping. Apparently, it worked pretty well, if you had the right people and were disciplined about the process.

Then Agile was applied to all strategic planning,  innovation and improvement processes.

The words behind VUCA and the words behind Agile entered the business lexicon; the discipline behind the processes not so much.

In a strategy discussion with a senior leadership team on leader kept using the acronym VUCA. This raised everyone’s anxiety level to the point that I finally showed the Bennett and Lemoine diagram and we began to work on what was of most concern, the uncertainty of a new technology introduced by a competitor.

I was in the middle of a continuous improvement project at a diversified financial services company when one division rejected the methodology we were working with. “Agile is much faster,” they asserted.

After three months and a sizeable investment they rejoined the corporate improvement initiative.

“We tried to do everything at once. Nobody was defining what needed to be improved.”

In change be suspicious of “silver bullets” or “quick fixes.” Details are important.

“Time, oh sweet, sweet time”

Another important factor is the time horizon of the change. This is critical when examining cost. There is initial cost, maintenance, and other life cycle costs. A friend is on the board of a townhouse condominium complex. He complained to me recently.

“These places were built in the1980s. It is time for new asphalt shingle roofs. We have two bids one is for 50% more than the other. Everyone wants to go the cheaper contractor. But it’s apples and oranges. The first guy is going to put a course of shingles right over the top of the shingles that are there. The more expensive roofer is going to take off the old shingles, remove the vapor barrier, check the roof sheathing and replace plywood where he has to replace the old tarpaper vapor barrier with newer material before replacing the shingles.”

“Do they last the same amount of time?” I asked.

“Well, the shingles have the same life 20- 30 years. The issue is the vapor barrier. If the tarpaper holds; it’s fine. If not with two layers of shingles you are not going to know where the leak is coming from and with the extra layer of shingles some of the runoff may run across the skylights. That probably won’t cause a problem for five years or so, but if or when it does it will be expensive.”

“Seems like a case of initial cost vs. lifetime cost” I said “Do you have the money to do it right for the long term.

“Yeah, but it is just hard for some to spend 50% more now.”

Some might make the argument that taking the low cost fix is the better deal because many people will move before any maintenance issues arise. Many businesses have short term time horizons, because in a VUCA world, who knows what will happen. But here is a case where it seems like we do know that if we take the short term approach we will be in for some long term problems. What is the cost of unintended consequences? How many roof leaks and leaky skylights would it take for the quick fix to be more expensive?

In times of change consider decisions from multiple planning horizons.

You may decide to go with a quick fix, but at least you will act with awareness of the long term.

“Jeanne is going on maternity leave, can we hire a temp to fill in for three months?”

“We have no budget for that, Paul. You and your team can pick up the slack for three months.”

The team did pick up the slack. Mostly the manager, Paul, picked it up. Unfortunately Jeanne had a difficult delivery and didn’t come back for five months and when she came back it was part-time at first, but she did come back and the team did the additional work without the cost of a temp. However, within a year, Paul and two team members left for other jobs. No one looked at the budget strain of hiring three new staff and rebuilding cohesiveness in the new team.

Sometimes you have to do a quick fix. Sometimes you fix something with duct tape. But remember where the quick fix is and have a plan to fix the consequences. Duct tape doesn’t last forever and using it on people isn’t recommended.

The quick fix’ll git you ef you don’t watch out.

 

The title of this post comes from combining the title from the book “Beyond the Quick Fix” by Dr. Ralph Kilmann and the Hoosier poet James Whitcomb Riley’s poem, “Little Orphant Annie,” which my parents used to read to us every Halloween.  Annie told “witch stories” with the moral “The Gobble ‘uns’ll git you ef you don’t watch out,”

“If you can’t measure it you can’t manage it.”

By the time you lead change you’ve been slapped sick with this cliché.

Many change leaders are managers, responsible for KPIs (key performance indicators). The phrase key performance indicators was coined by Art Schneiderman of Analog Devices in 1987, and popularized by David Norton and Robert Kaplan in their writings about the Balanced Scorecard.

The idea was a few simple measures across four perspectives:  financial, customer, internal operations, and innovation and learning. The intent was to encourage more metrics than just financial ones to operate the business in an integrated way. Unfortunately this has morphed into measuring everything all the time. Somehow the defining word ‘key” was lost, but now managers cringe at the letters K-P-I.

When managers hear about change metrics, the idea of being held accountable for more KPIs may not fall on receptive ears. Most organizations should reduce the number of metrics rather than increase them.

Screw-up: piling on change KPIs without rationalizing other metrics.

Measuring change – Don’t rely on single lagging metrics

However, change must be measured; otherwise how would you know that you’d changed. Whether a business is improving,  or innovating, there is a before and after, a delta.

Managers often talk about a single number, sales of $x, numbers of products shipped, or people trained. Even when managers talk in terms of ratios like return on capital employed (ROCE or earnings before interest and tax/ total assets-current liabilities ) or days sales in Inventory (DSI -products in inventory /daily product sales) it is a single number.

If you watch change leaders they talk in terms of two numbers, the delta. “$X sales last month vs. $Y sales this month and our DSI has moved from ninety days to fifty four days. Production isn’t keeping pace with the increase in sales and if we don’t improve we may face a stockout within six months.”

Successful change leaders talk trends. They understand that change is volatile, sometimes “two steps forward, one step back,” or even sometimes “one step forward two steps back,” but they keep focused on forward progress using overall trends to help.

Screw-up: focusing on a single metric at one point in time, “Profit is down!”

Leading vs Lagging

Profit is a lagging measure. It measures past revenue minus past cost. Many of the common metrics used by managers are lagging i.e., they measure things that occurred in the past that one cannot have an impact on. Theoretically managers manage in a steady-state environment so yesterday’s sales, yesterday’s operations, even last quarter’s  net employee attrition numbers or employees trained last year are relevant because this years are likely to be the same.

However when you are making change you want to know things before they happen not after. So it is better to know the new sales calls and the sales cycle, and call conversion rate trends. It is better to know scheduled deliveries of critical parts and planned maintenance or anything that might hinder your products-shipped goal. It is better to know why people are leaving, rather than last quarter’s metrics on net attrition.

Screw-up: waiting for lagging metrics; change leaders need leading metrics, indicators of what will happen before it has already happened. “You mean forecasting? Forecasts are never accurate.”

Leading Metrics Mistakes – The Perils of Forecasting

Forecasting is hard; It is measuring the future. Look at the trouble trained economists have at predicting market growth or recession. Look at how often marketing and sales people miss quotas. Even something that might look simple -a manufacturing production target: There’s materials in inventory or scheduled deliveries. You have labor scheduled. You know machine throughput. What could go wrong? Well, lots actually.

The production target is a future output, a results metric, a forecast. Manufacturing is a process to deliver that result.

A process is:

There are input measures materials in inventory waste, materials delivery schedule compliance, labor schedule compliance, etc.

Forecasting Screw-ups:

• Not foreseeing problems with inputs – e.g., high materials waste or missed deliveries
• Misunderestimating timing and flow of activities, g., machine breakdowns, strike
• Overestimating outputs – high inspection failure or output waste.

Even very experienced people have trouble with forecasting. Right now the Federal Reserve Bank chairman is wrestling whether to raise interest rates again to slow the economy and lower inflation without causing a recession. They key question is how much lag time is there between higher interest rates (an input) and lower inflation an output and a lagging indicator.

If experienced people have difficulty forecasting what do inexperienced people have? Excel.

The first time I saw this phenomenon the spreadsheet was Visicalc, then Lotus123 and now I have seen it multiple times using Excel.

An analyst, an innovation team lead, a student entrepreneur plugs much known data about new product costs into a spreadsheet, then reverse engineers the revenue line to show the revenue needed to fund the cost line. They end up with a graph that looks something like this:

They have no reasoning for why between year four and five; they just know it needs to if the project is to be economic.

Then something strange happens. Because it is in a spreadsheet and spreadsheets are usually used to track known data, lagging metrics that have already happened, they begin to believe and become committed to the projections in the spreadsheet. It was often my job to ask inconvenient questions. “Why does revenue jump between years four and five? What is your action plan to ensure that happens?”

“Er. . . ?”

Whereupon I was known to circle the jump and say “I see. . .  between years four and five a miracle happens.” I then label it: Excel enabled wishful thinking.

Screw-up: Having no plans to achieve a forecast and no contingencies if things go wrong.

Unverified Proxy Measures

“You can’t measure that!”

There are things that defy measurement. How would you measure when a product will generate enough word of mouth promotion that sales jump exponentially? How do you measure the difference between run of the mill “satisfied” customer and “delighted” customers who tell everyone they know about your product.

There have been several attempts at this. In the early days of the Internet there were some people whose opinion was such that it could generate word of mouth “virality.” Then using “influencers” became a strategy; now everyone is an influencer.

In 2003 Fredrick Reichheld created the Net Promoter Score (NPS) based upon the answer to the question, “How likely are you to recommend?” It was a “proxy measure” for customer delight.

It was a good proxy, but now many are sick of surveys and fewer people answer them. So is this a still a good proxy for the many companies who adopted NPS.?

This is one of the issues with proxy measures and in fact many leading indicators. They may be created by detailed correlation analysis, but once adopted they become gospel. No one checks the analysis.

When I moved to Pittsburgh in 1981 the office took me up on Mt. Washington to watch the Jones & Laughlin steel mills belching fire into the night. Three years later those mills were gone. Fully integrated mills were replaced by “mini-mills” that made steel from scrap sold through “service centers” that originally sold mill surplus production to small buyers.

I asked a friend who worked with marketing data at US Steel how this trend went unnoticed.

“We always thought of mini-mills as children’s toys and the service centers were just places we dumped stuff we couldn’t sell so we totally missed when their sales volumes ballooned through selling foreign steel and mini output. The biggest problem was that we defined the US market total steel consumption with a proxy measure. We divided the production of the Homestead Works by .12 because every year since before World Ward II, Homestead always had 12% of the market. Then suddenly they didn’t, but we didn’t know that.”

Screw-up: Not consistently monitoring and verifying proxy metrics. Even the most researched leading indicators can change.

Unintended Consequences

“Be careful. What you measure people will do.” This is a lesson we learn repeatedly in business.

In consulting I worked some in call centers, big rooms of gray carpet baffled cubicle where headset wearing customer service representatives (CSRs) CSRs sat staring at computer screens with customer account information. They were usually two or three shift operations with computerized call routing and online metrics on the CSR’s screens.,

CSRs were measured on many metrics and there was often metric conflict. One CSR confided,

“I’m told we want to solve the customer’s problem, but I’m measured and paid bonus based upon average call handle time. So if I get an unhappy customer who’s going to take a while, I disconnect while I’m talking. Then I pick up the next call in the queue. If they call back it’s someone else’s CHT, not mine.”

Screw-up: not examining metrics looking for conflicts and unintended consequences.

Avoiding Change Metric Screw-ups

All metrics are subject to errors. We can have too many metrics, or conflicting metrics or proxy measures that no longer predict outcomes. We can measure things that by measuring produce unwanted behavior or results. In times of change these mistakes may mean the difference between success or failure of the change. Here are some ideas for avoiding these mistakes:

• If you create a new metric, get rid of one.

• Find a “glass half empty person, (you know, the pain in the neck who always says stuff like “let’s look at the worst case scenario”). Have that person look at your metric for potential problems.

• Look at leading indicators and proxy metrics, but constantly revaluate them to be sure they still work.

• Beware of forecasts, test them for action plans to get there.

• Create action for when a metric shows something isn’t working.

And remember to share data and encourage people. Change is about more than numbers. Behind every changed number is a person or people doing more of something, less of something, or something differently.