How we work
Before changing the work, we need to understand it.
DeepWork analyzes how work really gets done inside an organization to identify where capacity goes, what creates friction and what should change.
We don't start from a tool, a technology or a predefined solution. We start from the work as it actually happens.
Real work rarely matches the designed process.
A procedure may list five steps. In practice there may be eight.
A task may look simple until you find out it means checking a spreadsheet, sending a WhatsApp message, waiting for an approval and re-entering information in another system.
A tool may exist, but the team may have built parallel processes just to get work done.
That's why we don't only analyze procedures. We analyze how work really happens.
We don't want people to work more. We want less work to be wasted.
As organizations grow, they pile up tasks, checks, meetings, documents, tools, approvals and ways of coordinating.
Many were necessary when they appeared. Some still are. Others remain simply because nobody asked again why they exist.
DeepWork questions the work before trying to make it more efficient.
The analysis process
From observing the work to deciding what to change.
-
01
Observe
What is really happening?
We identify the tasks, processes, tools, interactions and ways of coordinating in day-to-day work. We aren't looking for a solution yet: first we need to understand the operational reality.
-
02
Understand
Why does the work happen this way?
An approval may respond to a risk. A parallel spreadsheet may make up for a system limitation. A repeated question may exist because the information isn't where it's needed.
Before eliminating an activity, we need to understand what purpose it serves.
-
03
Quantify
How big is the problem?
When there is enough evidence, we translate friction into observable variables:
- Time
- Frequency
- Volume
- Steps
- Interactions
- Waiting time
- Rework
- Errors
- Dependencies
- Manual work
- Potentially recoverable capacity
Not everything can or should become hours or money. We quantify only when there is a reasonable basis.
-
04
Prioritize
What deserves to change first?
One activity may take a lot of time and be very hard to change. Another may need a minimal change and affect dozens of people every day. We assess:
- Impact
- Frequency
- Reach
- Complexity
- Dependencies
- Risk
- Investment
- Potentially recoverable capacity
The goal isn't more recommendations. It's knowing where to act first.
-
05
Intervene
What should change?
Once the problem is understood and prioritized, we apply the DeepWork framework as an order of decision. Not every opportunity goes through all four stages.
-
06
Verify
Did the change deliver the expected result?
When we implement an improvement and there is enough information, we compare how things work afterwards with the starting point: time, steps, manual work, errors, rework, waiting time or capacity.
Implementing something isn't the same as proving the work improved.
The DeepWork framework
Four questions before deciding on a solution.
Should this work exist?
Before improving an activity, we question whether it is needed. It could be:
- A meeting
- A report
- A data entry
- An approval
- A check
- A document
- A process step
- A recurring question
If an activity doesn't create enough value, making it more efficient doesn't solve the problem.
If it must exist, does it need to be this complex?
We look to reduce:
- Steps
- Interactions
- Approvals
- Exceptions
- Tools
- Information handoffs
- Dependencies between people or departments
So the necessary work takes less effort to produce the same result.
If it must be done, does a person need to do it?
We analyze repetitive, predictable or rule-based activities that can be handled, fully or partly, with:
- Existing tools
- Integrations
- Automations
- Artificial intelligence
- Specific development, when justified
Technology comes in after understanding and simplifying the work.
Where does human work create the most value?
Not everything should be automated. Decisions, relationships, negotiation, analysis, creativity and complex situations may need human judgment. We reduce the surrounding work and interruptions that compete with those activities.
We don't want to automate people. We want to free up human capacity for work where it truly adds value.
One problem, different solutions
We don't know the solution before analyzing the problem.
“We waste a lot of time looking for documents.”
The solution could be
- Reorganizing how information is stored
- Removing duplicates
- Defining a single source of truth
- Changing permissions and responsibilities
- Making better use of the existing tool
- Adding smart search or another technology
“Our salespeople have little time to sell.”
The solution could be
- Dropping reports
- Simplifying admin work
- Redistributing tasks
- Changing internal processes
- Integrating systems
- Automating repetitive activities
“We have too many meetings.”
The solution could be
- Changing how decisions are made
- Changing how information flows
- Reviewing who needs to attend
- It may not need any technology
The problem defines the solution. Not the other way around.
What we observe
Work is a system.
-
01
People
Roles, responsibilities, dependencies and the interactions needed to move forward.
-
02
Processes
Sequences, decisions, exceptions, checks and waiting time.
-
03
Spaces and operations
Walking routes, movement, waiting, preparation and the execution of physical work.
-
04
Information
How it is created, stored, found, shared and transformed.
-
05
Technology
Systems, tools, integrations and the manual tasks that exist around them.
-
06
Communication
Meetings, messaging, email, questions and interruptions.
-
07
Capacity
Where time goes and which activities compete for it.
Friction tends to appear between these elements, not inside just one.
Sources of evidence
There isn't just one way to understand work.
Depending on the problem and the scope, we can combine:
- Interviews
- Observation
- Task analysis
- Process mapping
- Time measurement
- Existing documentation
- Operational data
- Calendars and meetings
- Systems and tools
- Workflows
- Available records
- Analysis of information and communication
Not all of them are needed in every project.
We choose the evidence based on the question we need to answer.
What we don't do
Analyzing work doesn't mean monitoring people.
We don't assume a productivity problem is the result of individual performance. A person may be working very hard inside a needlessly complex system. That's why we look for problems in:
- Processes
- Information
- Tools
- Coordination
- Decisions
- How work is distributed
- Operational design
The goal is to improve the system of work, not to rank employees.
From friction to capacity
Not every busy hour is capacity well used.
Picture a team with 1,600 hours available per month. After the analysis, we find:
1,600 h / month
That doesn't mean 370 hours can be recovered automatically. Each activity is analyzed: some can be eliminated, others simplified, others automated, and some will have to stay.
- Necessary work
- 1,050 h
- Coordination
- 180 h
- Repetitive manual tasks
- 120 h
- Rework
- 95 h
- Searching for and rebuilding information
- 80 h
- Other friction
- 75 h
Potentially recoverable capacity
The share of capacity used today that offers a reasonable opportunity for reduction through changes to the system of work.
DeepWork isn't a recipe
Different companies can have the same symptom for completely different reasons.
Two organizations may say: “We have too many meetings.”
In one, the problem may be a lack of information.
In another, overly centralized decisions.
In another, unclear responsibilities.
In another, systems that don't show the status of an operation.
Cutting meetings across the board could make the problem worse.
The methodology seeks to understand the cause before changing the symptom.
Frequently asked questions
About the methodology.
How do you decide what to eliminate, simplify or automate?
First we observe how the work happens and understand why it works that way. When there is evidence, we quantify the friction and prioritize by impact, frequency, reach, complexity, risk and investment. Only then do we apply the framework: Eliminate, Simplify, Automate and Focus, in that order.
Do you always apply all four stages?
No. The framework is an order of decision: one activity may be eliminated outright, another only needs simplifying and another is worth automating. What doesn't change is the order of the questions.
Is everything measured in hours or money?
No. We quantify only when there is a reasonable basis for it. Some friction is better described in steps, waiting time, errors or dependencies, and we don't use universal productivity percentages.
Do you follow a standard methodology?
We combine well-known work analysis techniques, such as interviews, observation, process mapping and time measurement, depending on the question we need to answer. We don't apply a recipe: two companies with the same symptom may need different solutions.
Do you evaluate each person's performance?
No. We look for problems in processes, information, tools, coordination, decisions and how work is distributed. The goal is to improve the system of work, not to rank employees.
DeepWork methodology
Before deciding what to change,
let's understand what is using up
the capacity.
We analyze the work, identify friction and prioritize opportunities to eliminate, simplify, automate and protect the highest-value work.