Transparency builds trust, including the hard parts

In the 1970s, the psychologists Sidney Rosen and Abraham Tesser documented something they called the MUM effect: people systematically avoid delivering bad news, even when they have nothing to gain from silence and the recipient badly needs to know. It isn't strategic dishonesty. It's discomfort. We hold the message rather than hold the awkward moment.

Now install that instinct inside a software project, add hierarchy and a billing relationship, and you get the industry's most familiar artifact: the watermelon project, green on the outside, red on the inside. Researchers who study troubled IT projects, notably Mark Keil and colleagues, have shown how consistently status gets shaded on its way up the chain. Each layer softens the report a little. No one outright lies. And the executive sponsor learns the truth on the day it can no longer be fixed. Most project disasters aren't information failures. The information existed. It just never traveled.

Which is why our eighth commitment isn't really about honesty as a virtue. It's about building structures that make the truth travel, especially when it's uncomfortable.

Placement is policy

The MUM effect doesn't yield to good intentions, because it operates precisely when intentions are good. It yields to structure. So: bad news leads the weekly report. First section, above the accomplishments, every week. This sounds cosmetic and is anything but. If bad news is allowed to sink to the bottom, the MUM effect will sink it, week after week, one soft framing at a time. Making it lead turns candor from a personal act of courage into a formatting requirement. The engagement lead doesn't have to decide whether this is the week to be brave; the template already decided.

The same structural logic runs through the rest of the commitment. We walk clients through pricing and engagement structure before anyone signs, because a surprise discovered in the commercial mechanics in month two poisons every status report that follows. Once a client wonders what else wasn't mentioned, everything gets re-read in that light. And we show our methods, including the approaches that didn't pan out, because a vendor who only ever reports successes is describing a probability distribution that doesn't exist. Sophisticated clients know it.

Honest numbers come with ranges

Nowhere does fake certainty do more damage than estimation. Barry Boehm's research, later popularized by Steve McConnell as the cone of uncertainty, found that estimates made at the very start of a project can be off by a factor of four in either direction, narrowing only as real decisions retire real unknowns. That isn't an indictment of estimators. It's the actual shape of knowledge at the start of a project.

So a single-number estimate, "this will take twelve weeks," is a confidence trick, even when it's an accidental one. It reports precision that cannot exist yet. Our estimates come with a confidence range, and the range narrows as the cone says it should: early on, "ten to eighteen weeks, and here are the three unknowns driving the spread." A client can plan around an honest range. What nobody can plan around is a precise number that was always fiction, and that, per the MUM effect, gets discovered to be fiction at the worst possible moment.

The cost, and the return

Transparency has a price, and pretending otherwise would itself be a transparency failure. Leading with bad news makes some weeks tense. Honest ranges have lost us deals to competitors quoting confident fictions. Showing failed approaches means clients occasionally watch us be wrong in real time.

Here's the return. Clients don't actually leave over problems. Years of watching engagements succeed and fail has convinced me of this. They leave over surprises: the gap between what they were told and what turned out to be true, because that gap is where trust dies. A hard truth delivered early, with a plan attached, almost always deepens the relationship. The same truth discovered late, after a month of green reports, ends it.

The Sigao take

We communicate plainly, share what matters, and keep the commitments we make, and we've learned not to leave any of that to personality or courage-of-the-moment. The MUM effect is human and permanent; structure beats it where willpower doesn't. Bad news first. Prices before signatures. Ranges instead of false precision. Methods shown, failures included. None of it is comfortable, and all of it is why a client can take our green status report at face value: they've seen what our red one looks like.

Sources

• Sidney Rosen and Abraham Tesser, "On Reluctance to Communicate Undesirable Information: The MUM Effect", Sociometry, 1970.
• H. Jeff Smith and Mark Keil, "The reluctance to report bad news on troubled software projects: a theoretical model", Information Systems Journal, 2003.
• Steve McConnell / Construx, "Software Development's Cone of Uncertainty": Boehm's original data, popularized in Software Estimation (2006).