Be correct first, then be correct fast
It feels like a trade-off in your first months. It is a sequence, it only runs in one direction, and getting the order wrong will cost you a year.
Clients pay somewhere between hundreds and thousands per consultant per day. A normal project runs into the millions. For that money they expect flawless work, and they are right to.
Almost every consultant has the same story. An important read-out, months of good work behind it, and a client spots a typo in a footnote on slide 183. They point it out in the meeting. Five minutes of meeting time gone, and, much worse, a small crack appears in the credibility of everything else on the page. Quality equals credibility, and it works in both directions.
What errors actually cost
The bigger errors do more than dent your credibility. Wrong data, a faulty calculation or a conclusion that does not hold can cost weeks of rework, damage the client relationship permanently, or lead to a genuinely bad decision.
Here is one I still think about. A manager in our Swiss office ran a due diligence where our conclusion came out far more negative than the fund’s initial hypothesis, which had been built on the investment bank’s numbers. That triggered a week of intense discussions with every senior stakeholder about how two teams could land so far apart. Extra nights for the consultant and associate rebuilding the model, the senior manager checking calculations line by line, and the fund’s own team trying to reconcile it all.
The cause was one transposed growth rate. 7.3% instead of 3.7%, somewhere in sheet five of the bank’s file. Hundreds of thousands in additional fees, a lot of stress on all sides, and real damage to trust, all from two digits in the wrong order. Although it is worth saying: finding it prevented a multi-million investment into a company that was never going to hit its targets.
Inside your own team it plays out more quietly. Each error you make now shows up later as extra work, usually at the worst possible moment. Call it efficiency debt. Your supervisor’s confidence drops a notch, they start checking more of your work, and the checking costs everyone time you no longer have.
The trap almost everyone falls into first
When you start, you will feel slow next to everyone around you. The natural response is to go faster, which costs quality, which gets you reprimanded, which feels like an impossible standard. Be fast. Also be perfect. Pick both.
On my first case at Bain, I was obsessed with speed. I wanted to impress. My slides came back covered in stickers, comments and corrections on every page. I turned them around in under two hours and felt proud of it. They came back again. I turned them around again. I was convinced that what mattered was being fast, that the consultant would check my work anyway, and that a few typos and misalignments were somebody else’s problem. What counted was substance.
After the third round, the consultant pulled me aside.
“You are really fast. But I can never trust your output if there are this many errors in it. Take some of the speed away and get the slides right. How else would I ever trust you to work directly with the client? And how else can you learn to review someone else’s work when the time comes?”
That conversation shifted something fundamental. Speed without accuracy is not speed. It is rework, and rework costs more than getting it right the first time. Worse, it moves work from me to the consultant, who is more expensive and more useful elsewhere.
The mechanism, stated plainly
Every error you send upward creates a correction loop. While those loops exist, your supervisor double-checks everything you produce, which caps how fast the team can move regardless of how fast you type. Once they trust that your slides, models and emails are clean, the double-checking stops.
That is the moment you get faster. It is also the moment you get more responsibility, more direct client work, and fewer review rounds. The sequence is fixed:
1. Build trust through accuracy.
2. Convert that trust into speed using every technique in this series.
3. Convert the speed into scope, because people give more interesting work to those who return it clean.
Trying to run it in the other order produces an associate who is quick, unreliable, and permanently supervised.
A philosophy rather than a target
You will never reach zero errors, and chasing perfection is the wrong goal. What you build instead is a disciplined approach: systems, habits and checks in your workflow that make errors rare and, when they happen, catchable before they leave the team. A manager I worked with used to say it better than I do: being error-free is not a method, it is a way of life.
Which has one practical implication that people skip. Always plan time for review. Roughly 5 to 10% of a work package once you are practised, 10 to 20% while you are learning. If you have thirty minutes for a task, three to five of them belong to the review. Do this even under time pressure, and especially under time pressure, because that is when the errors are.
START THIS WEEK
1. On your next three deliverables, book the review time in the calendar before you start the work. Not after. Treat it as part of the task rather than as something you do if there is time left.
2. Count your correction loops for one week. How many pieces of work came back for a second round? That number is your efficiency debt, and it is the number that decides how closely you are supervised.
3. Ask your supervisor one question at your next check-in: “What would have to be true for you to stop double-checking my output?” The answer is your development plan for the next month.
Accuracy unlocks trust, and trust unlocks speed. It does not work in the other order.