Method

Twelve points of inefficiency, five months, 258,000 deliveries.

In 2019 I wasn't looking at investment portfolios. I was looking at stalled cargo in a warehouse, with the board demanding results in the middle of a merger. What I did there became my MBA thesis at USP/Esalq (the University of São Paulo). It's the same reasoning I use today when someone opens up their financial life in front of me.

  • 19% → 7%Monthly inefficiency, March to August 2019
  • 258,000Deliveries analyzed in Greater São Paulo
  • 6,511 → 2,120Stalled loads per month, a 67% drop
  • 28 → 7People in the control area after the redesign

Monthly delivery inefficiency — January to August 2019

Stalled loads as a share of total delivery attempts, Greater São Paulo.

Monthly delivery inefficiency, January to August 2019 Percentage of stalled loads over total delivery attempts. It rises from 15% in January to a peak of 19% in March, when the project was applied, and falls to 7% in August. The company's target was 10%. 0% 5% 10% 15% 20% After the project Jan/2019 — 15% inefficiency · 4,961 stalled loads in 32,830 delivery attempts Jan Feb/2019 — 17% inefficiency · 5,136 stalled loads in 30,424 delivery attempts Feb Mar/2019 — 19% inefficiency · 6,511 stalled loads in 33,777 delivery attempts Mar Apr/2019 — 13% inefficiency · 4,767 stalled loads in 35,549 delivery attempts Apr May/2019 — 11% inefficiency · 3,838 stalled loads in 35,167 delivery attempts May Jun/2019 — 12% inefficiency · 3,443 stalled loads in 28,743 delivery attempts Jun Jul/2019 — 8% inefficiency · 2,610 stalled loads in 32,072 delivery attempts Jul Aug/2019 — 7% inefficiency · 2,120 stalled loads in 30,261 delivery attempts Aug 19% 7% Company target: 10%
Above target Below target Company target (10%)
View the data as a table
MonthDelivery attemptsStalled loadsInefficiency
January32,8304,96115%
February30,4245,13617%
March33,7776,51119%
April35,5494,76713%
May35,1673,83811%
June28,7433,44312%
July32,0722,6108%
August30,2612,1207%

Source: operational data extracted from the carrier’s TMS (transport management system), compiled in the thesis “Reestruturação da área de Controle de Operações para implantação de Projeto de Redução de Ineficiência através do Gerenciamento de Cargas Paradas” (Restructuring of the Operations Control Area to Implement an Inefficiency Reduction Project through Stalled Load Management), MBA in Project Management, USP/Esalq (Pecege), 2020.

What was happening

  • The company had been formed by buying four Brazilian carriers. Four cultures, four ways of working, one system. Nobody had stopped to redesign the process after the merger.
  • The internal inefficiency target was 10%. In January we were at 15%. In March, 19%: almost double the target, and the curve still rising.
  • Twenty-eight people in the control area, each with a scope so broad that handling stalled loads — exactly what generated customer complaints — became the last task of the day. When there was any day left.
  • Control, resolution and the warehouse traded emails instead of talking. Each area became an island, and the answer that would save a delivery arrived after its deadline.

What I did

I asked before deciding

Before touching the org chart, I ran a 26-question survey with the eleven people who did the work every day: what blocks them, what trips over what, where communication fails. The first meeting was tense and full of cross-accusations. It was also the most useful one: the map of problems came out of it, not from a consultancy.

I broke big tasks into small ones

A scope that is too wide is not a sign of a versatile team. It’s a sign of poorly defined priorities. I redesigned the roles into smaller blocks: one person solely for scheduling, four for problem-free loads, one analyst dedicated to indicators. Each person ended up with a slice that fit into a workday.

I created a team that only handled what went wrong

Six people left control and formed a cell dedicated exclusively to stalled loads, with no other task competing for attention. And that cell moved physically next to the warehouse: they could now see, through the window, the truck coming back with the load that didn’t get in. Distance between those who decide and those who execute costs money. There, you could see the cost in the indicator.

I took the middleman out of the customer conversation

Before, every failed delivery went through the sales team before reaching the shipper. I cut the middleman: the resolution cell started talking directly to whoever shipped the load. Sales went back to selling, resolution gained speed, and the goods stopped aging in the warehouse waiting for an email to be answered.

I measured every day, not every month

The service-level indicator began to be calculated and read daily by the team and management, instead of being closed at the end of the month, when nothing can be fixed anymore. That is what let us see the curve turn in April, three weeks after applying the project, and not in July.

What this has to do with your money

The question I asked in that warehouse was: where does the process get stuck, how much does that cost per month, and what is the single change that moves the number. It’s the same question I ask when a business owner opens up their structure in front of me.

A company that exports and carries expensive working capital in reais has a mismatch, not an investment problem. A family with seven products across four institutions and no consolidated view has a control problem, not a returns problem. In both cases, changing the product before understanding the flow is like switching suppliers without looking at the process: sometimes it improves things, almost never does it solve them.

That is why I came to this side. The rigor is the same. What changed was the object.

What this case is not

This is a study of logistics operations, not of investment. It does not represent, suggest or project any financial result on investments, and no transposition is possible between an operational indicator and the returns of a portfolio. It is here for one reason only: to show how I work through a problem before offering an opinion on it. Investments involve risk, and past performance does not guarantee future returns.

Bring me your number, not your doubt

If you have a business or a body of wealth with scattered pieces, start with what can be measured: what comes in, what goes out and what sits idle. In a no-cost diagnostic conversation, we build that map together.