A Six Minute Read.
As director of the Project Management Office, Cyrus supervised a dozen PMs (each with a caseload of ten or so projects), but was not involved in the sales process. As such, he didn’t see proposals or contracts until after they were signed by the client and returned to Natalie, the accounts coordinator. (His PMs often helped the business analysts during the sales phase, and even participated in client visits.) Once Natalie received a signed agreement, she would copy key information into a Project Request Form and forward it to Cyrus, who would review the request form and ultimately assign the project to one of his PMs.
Cyrus usually could get the project off his desk and on a PM’s in less than a week, so the sales team built that into their estimates with the client. I asked Cyrus how much time he spent reviewing each project: some required consultation with the BA to understand the scope, but for most, “just a glance will do”.
I asked Natalie if she knew how Cyrus decided on project assignment, and she told me that certain technologies always go to specific PMs, and clients were all pre-assigned to certain PMs. In less than ten minutes, we drew up a table showing which types of projects, and which clients, went to which PMs. I grabbed the most recent 100 project request forms, and based on the information in the header, correctly guessed how 93 of those projects were assigned. (Four were swip-swapped by the PMs themselves; three were reassigned around one PM’s maternity leave).
I asked Cyrus why he spent so much time reviewing the project requests before assigning them, when it was already clear who would ultimately get them. He told me, “I like to know what my people are working on.”
Of all the possible disconnects in business processes, some of the most frustrating – and insidious – are filed under “Bureaucracy”. Many characteristics used in designing world-class processes focus specifically on firewalling potential bureaucratic behavior. When targeting process improvement efforts, “eliminating bureaucracy” – that is, getting rid of redundant (and repetitive) procedures and unnecessary tasks – is almost always the first place to start.
It’s easy to pick on “government bureaucrats” – just recently, White House Aide DJ Gribbin conducted his own “brownpaper” review with President Trump, highlighting the byzantine path to getting a highway built with federal funds. Such soul-sucking is not limited to civil servants. In most organizations, the bureaucracy – and its hidden costs – is so pervasive that even if no other process improvements are made, eliminating these behaviors will reap plentiful, and ongoing, rewards.
Two of the most common, and commonly grating, bureaucratic activities are “approvals” and “inspections”. Neither offer much value, and eliminating them significantly improves efficiencies – and effectiveness.
“Approvals” come in many forms: “Let me read that before you send it out”; “All requisitions must be accompanied by a supervisor’s signature”; “No one can use the conference room without Stacy’s permission.” The arguments in favor almost write themselves: “I don’t want to be blindsided!”, “It’s my job to protect the budget!!”, “If no one watches it will become chaos!!!”
Make no mistake, each example is a tragic waste of time, money, and human spirit. In most cases, an “approval process” indicates a fundamental lack of trust: in others’ abilities, judgment, or maturity. (Sometimes, like Cyrus, it’s because we just “like to know” things. That’s a poor excuse: there are lots things I like to do that I don’t at work.)
Often we build approvals into the process because we can’t (or won’t) properly delegate the work. It’s “easier” to have someone else start, bring it to us for our two cents, and then kick it back to them for finishing. That’s not “approval”, that’s insecurity.
Sending things back for revision not only adds delays (time is money), but also incurs rework costs.
How to Fix:
Ask: How many times have you not approved something? Often, the answer is “never.” If that’s the case, it’s not an “approval” step, it’s a “perusal” step, and can be easily eliminated. (The argument: “But it only takes me a moment to read!” actually underscores how non-value-adding it is.)
The next step in fixing this is to review rejected items, and study the reasons they were not approved. Pareto the most significant, and then train the staff to not make those mistakes again. (Hint: Your employees understand how budgets work. Don’t use a wasteful “approval process” in lieu of enabling them to make good decisions.)
One special case I see frequently is “multiple” signatures. I came close to blows with one client who wanted to change an approval process from one-required-signature to two, in the aftermath of a systemic failure. Rather than deal with the reasons for the breakdown, the “easier” solution was to add a second set of eyes – and another round of delays.
Inspections are essentially after-the-fact approvals. All the real work is complete, but someone needs to “make sure it’s right” before a customer sees it. This is different than a collaborative review in creative or one-of-a-kind endeavors (“Can I walk you through this presentation?”, “Read this – does this even make sense?”, “They’re giving us 24 whole dollars for this island – what am I missing here?”).
Most inspections – and this includes the bulk of the work done by the “Quality Department” – offer zero value. By design and definition, an inspector is merely making sure someone else did the job he was assigned.
How to Fix:
Easy to write, complicated to fulfill: Train the staff to do it right the first time. Then, dutifully work through each step of their process and find ways to eliminate the potential for error. Create a system that prevents incorrect pieces to move on in the process, and rejects incoming mistakes. Can’t figure out how to do that? Ask the people doing the job – they know. Whether you call it poka-yoke or simply good design, do it.
Even if approval and inspection processes don’t offer much value, they don’t offer much harm, do they?
Sure they do.
To appreciate the true cost of bureaucracy, two elements must be considered –the “duration” of each activity and its “time on task”. “Time on task” is fairly straightforward: How much time someone spends actually interacting with the process (e.g., the time Natalie spends cutting-and-pasting from the sales contract to the Project Request Form, the amount of time Cyrus spends reading the request form).
“Duration” is the time from the start of one task until the time the next task starts. So the time-on-task for Cyrus to read the Project Request Form and make an assignment may be six minutes; however, it is over six days duration from the time Natalie distributes it until Cyrus assigns the project. (And, remember, my cheat sheet is nearly perfect in making the same assignment, in no time at all).
You can calculate the cost of time-on-task (multiply time spent with labor rates) to determine if the activity is worth it. The duration presents a less-clear picture in terms of expense, but nicely frames the opportunity costs of how a single step – such as Cyrus’s “review” – consistently delays client work.
Once you know the impact, you can also make decisions about needs for revision, quality tolerances, and more. If you are looking for money to add to the training budget, it is probably locked up in here.
Think about this from a customer’s perspective. Approvals and Inspections slow down work, and add cost (which is passed on to the customer), and only serve internal needs. Unless government-mandated, or a true life-or-death checkpoint, this bureaucracy should be eliminated.