Admit mistakes
We are not perfect, no firm is. Best to say, oops we blew it, than to try to brush a mistake under the carpet. Clients are quite aware that there is a trade off between accuracy and time spent. That said, we should strive for 97.5% accuracy. Making a lot of mistakes in a short period of time hurts our credibility. So does arrogance which often manifests itself in statements about how smart we are (not) or how bullet proof our processes are (ditto). If you doubt this, think about Microsoft and its “trustworthy computing” initiative. Widely publicized, this initiative was supposed to reassure the public that it was “on top” of its security problem. Instead, it had the opposite effect, especially after it became known that the slammer worm had infiltrated Microsoft itself. The result? IT is taking a good long look at LINUX as a viable alternative to Microsoft Server.
Know the problem
Most of the time, errors creep in at later stages of a project. Instead of doing things at the last minute, pace yourself so that you can get the deliverables you need well done before they’re actually needed. This is the single best way to avoid making mistakes; by being proactive you are giving yourself the time you need to find mistakes and correct them.
Keep it simple
Yes our analyzes can get complex. Mistakes are like mosquitoes1; there is a tendency for them to breed if you let the complexity stand.
The best way to build simplicity into your work product is to breakdown your analysis into bite-size chunks. Explicitly think through how a sanity checker can check the results of each chunk. Use the chunks as building blocks and you’ll know that each piece of your analysis has been sanity checked.
Distinguish sense from nonsense
Some findings are anomalies of the data. Others are real. Distinguishing between them takes experience and the willingness to look at different data different ways. Good is having one set of facts or analytics to support your conclusions. Better is having two sets of facts or analytics. Best is having three different facts or analytics that converge. Convergence means you can be sure the inference you are drawing from the data is real and not an artifact.
Ask for help
It’s hard to find errors in a document or analysis that represents your personal sweat and tears. Someone who has not worked on the project needs to serve as the sanity checker — to proof the content and fact check documents for obvious errors in thinking, logic, and mathematics.
Learn to listen
Every person here at Firewhite has a valuable role to play in getting to a quality work product. Think about Marilisa Walski, one of our associates here at Firewhite and a member of the founding team. Marilisa is often involved in producing the final work product. In that capacity, she’s the “gold standard” of what can go out in terms of look and feel. Listen to her feedback. If she does not feel something is up to our usual quality standard, she CAN and WILL call for a “do over”.
Likewise, as the CEO I am the “gold standard” in terms of overall quality of the strategic thinking. This means I should see/sign off on every major work product BEFORE it goes out to the client and I reserve the right to ask you for a “do over” if the thinking isn’t up to snuff.
A “do over” is not something to take lightly; it isn’t a lot-o-fun and will invariably mean that you’re here late into the evening and/or through the night so as to meet the deadline we promised the client.
Foot all your calculations
Clients can and will make million-dollar decisions based on the work that we do. This means we must have mechanisms in place to sanity check critical calculations. Most of the time, when errors creep into our work product it happens due to the one of the following:
- a number got transposed
- Microsoft Office scrambled the data on its way from Excel into PowerPoint2
- a consultant didn’t take the time to foot their work with row and column totals
- a sanity checker didn’t check the data on the slide against the original source
Footnoting your work will help both you and the sanity checker assigned to your team zero in on the problems. The process is simple. Number each calculation and use footnotes to your document to describe exactly how you worked the numbers. The notes can be for internal use only or presented to the client – that’s up to you. Typically, we provide workbooks with most of our analysis, to enable the clients to look at the assumptions we made and the work we did to get to the result of interest. This sounds like a lot of work (and it is) but it will actually save you a lot of time if you have to adjust your numbers later.
Accept change as inevitable
Ideally, production isn’t where we should be catching the majority of our mistakes. We’d all like to identify mistakes BEFORE the final production of the presentation or other deliverable. That said, we’re a young firm and a team that is working very hard and wearing many hats. For that reason, it is important that you stay hands on through the final production process including bindery. Avoid the temptation to dump and run — when you think something is finished and hand it off to someone else, in the hopes that everything will be “all right.” Only rarely will that be the case. Which means you should stay and see the project through to the bitter end.
Learn to ask questions
Making a mistake won’t kill your career, at least at Firewhite. What will is failing to ask why a particular mistake happened. Codify what you learn in something we call a “mistakes memo”. A mistakes memo is a description of what went wrong, how you figured out it was wrong, and what to do about it for next time. Writing this down and putting it on the Extranet means that our collective learning is available to all. Without this there is a tendency to repeat our mistakes over and over again, as new people join the business and old people leave.
Remain calm.
The single best way to recover from a mistake is to handle it with grace, dignity, and aplomb. Slow down. Breath. Acknowledge and recognize the problem. Commit to correcting it with all due speed. If you’re the lead person on the engagement, take ownership of the mistake by using the “I” word as in “I blew it.” (Conversely, when being praised use the “us” word. “Thanks for that high praise. It means a lot to us, particularly to Alex and Chinh who really made this project shine.”)
Notes
1. If you missed this lecture in biology, mosquitoes breed in standing water.
2. A known problem (is it a bug or a feature? Only Bill Gates knows for sure) with Microsoft Office as reported in PC Magazine.
Originally published on Firewhite Consulting site, 5.03.