Post-Decision Confirmation Continuation and Reflection

The Client We Won Twice

We threw a dinner to celebrate the win. Turns out, the client was still deciding whether we'd won.

The mandate was signed, the press release drafted, the team reassigned to the next pursuit, standard operating procedure. What I didn't understand then is what the first ninety days looked like from the client's side of the table. Our champion had spent political capital getting us hired over a safer name. Every Monday she walked into a meeting with the people she'd overruled, and every small hiccup of a new engagement, the onboarding delay, the invoice formatted wrong, the deliverable that needed one more pass, was being entered into evidence in a trial we didn't know was running.

Nobody complained. That was the dangerous part. The relationship just cooled, degree by degree, until a comment reached us secondhand: "starting to wonder if we made the right call." Ninety days after winning the business, we were losing it, without a single complaint being filed.

We recovered, barely, and the recovery taught me the rule I've carried since: the signature doesn't end the sale. It moves the sale inside the client's building, where they conduct it without you. The research on what's happening in there, and what to do about it, fills the rest of this entry.

The Decision Doesn't Stay Decided

Leon Festinger's cognitive dissonance theory starts with a simple friction, beliefs and actions that don't match create discomfort, and people work to reduce it. Right after a meaningful decision, that tension takes on a familiar shape. Suddenly the rejected option looks better than it did yesterday, and the chosen one shows its seams. To calm the unease, the buyer begins editing the story, praising what they picked, diminishing what they didn't, and adjusting their memory of the tradeoffs until it feels coherent again. So the decision keeps getting made again, just internally, after the external commitment.

That is why the early period after a "yes" is so delicate. A first experience that fits the buyer's self-justifying story steadies everything. A first experience that doesn't fit doesn't always cause a complaint, it often causes a quiet, nagging contradiction, the kind that eventually turns into drift.

My champion's Monday meetings were Festinger's lab, running live. She needed material for the story she was telling herself and her colleagues, and we were sending invoices.

Satisfaction Lives in the Gap Between What Was Expected and What Shows Up

Richard Oliver gave durable language to this with expectation-confirmation theory. Satisfaction isn't a verdict on absolute quality, it's a comparison. The work can be objectively good and still disappoint if it lands below the mental line drawn beforehand. Surprise people on the upside and they're satisfied, miss the mark and they aren't, even when the baseline is respectable. What was promised, implied, or assumed before the decision becomes the ruler used afterward.

That is also the cleanest explanation for why overpromising is so corrosive. It doesn't just create a demanding client, it rigs the yardstick so that normal performance reads as failure.

Continuation Is Its Own Choice

Anol Bhattacherjee took the same structure and pushed it forward in time. Adoption, he argued, is not the same thing as continuance. People don't keep using something because they once started, they keep choosing it because early experience either confirmed expectations and produced satisfaction, or it didn't. Retention looks less like a single relationship and more like a chain of small renewals, each one a near-silent replay of the original question: is this still right for me?

Experience Doesn't Become Expertise Until Someone Reflects

From the provider's side, the post-decision window is also the point where capability is either sharpened or wasted. David Kolb's learning cycle makes the hinge obvious: experience, reflection, concept, experiment, then back again. Remove reflection and the experience doesn't mature into anything, it's only something that happened.

Donald Schön, writing about professional practice, made the same claim in a different register. Skill grows through reflection in the middle of the work and reflection afterward, not from technical knowledge alone. Outcomes don't teach by themselves. They sit there inert until someone does the interpretive labor that turns them into judgment.

Chris Argyris and Donald Schön added the distinction that makes reflection more than a checkbox. Single-loop learning corrects an error without touching the underlying assumptions, fix the tactic, keep the frame. Double-loop learning is harder, it questions the assumptions themselves: was the goal wrong, the frame mistaken, the belief behind the decision flawed? Most people and most organizations stay in the single loop because it feels productive without being threatening. Double-loop work is rarer precisely because it puts your earlier thinking on trial, and that discomfort is often where the real improvement is hiding.

Here's my confession on this one: every firm I've worked with runs postmortems on losses. Almost none run them on wins, which is how the wrong lessons get to live forever. A win confirms everything you did, including the three things you did badly and got away with.

Why This Belongs in Decision Architecture

Decision Architecture treats Outcome and Reflection as part of the engagement, not stray cleanup, and the research explains why that's the right move. Commitment closes the evaluation that leads up to a choice, but it opens a second evaluation that follows it. Buyers keep checking whether they were right, and that continuation question sits at the center of client experience, retention, and referrals. Festinger explains the urge to justify. Oliver and Bhattacherjee explain the mechanics of satisfaction and continuance, how they form, and why they shift.

For providers, the implication is plain. The post-decision phase isn't "service" stapled onto a sale. It is the moment trust is either strengthened or quietly eroded.

The reflection literature anchors the other side of the framework. Experience alone does not produce expertise, unreflected experience is just time passing with better branding. Kolb and Schön supply the mechanism. Argyris and Schön supply the depth. It's the same loop that creates the pattern recognition described elsewhere in this library. The expert isn't the person with the longest list of outcomes, it's the person who extracted judgment from those outcomes, especially the person willing to question the assumptions that produced them. That is where an outcome becomes judgment, and judgment is what the next decision is built on.

Artificial intelligence can do the measurement, not the reckoning. It can track results, compress feedback, and point out where performance missed expectations, the confirmation step that drives satisfaction and continuation. What it cannot do is the reflection that matters most. Double-loop learning requires living with the discomfort of having been wrong, owning the call, and putting your own assumptions under scrutiny, and a machine has no stake in that, no accountability, no dissonance to settle.

There's a quieter risk as well. A neat AI post-mortem can create the sensation of reflection without the strain of it, a polished illusion of understanding that flatters your past decisions. That's exactly how elapsed time masquerades as expertise. AI can tell you what happened. Only you can decide what it meant, and whether you were wrong. That reckoning is where expertise gets made, and it is the one step that cannot be handed off.

The Marketing Read

The signature doesn't end the sale. It moves the sale inside the client's building, where your champion conducts it without you, every week, in meetings you'll never see. Marketing that stops at the win is leaving the second sale unattended.

So treat the first ninety days as a campaign, aimed at one voter. Your champion overruled somebody to hire you, and Festinger says they're editing the story of that call daily. Feed the edit. The early-win memo they can forward without being asked. The thirty-day summary written for their boss rather than for your file. The small deliverable that lands early and clean, chosen for visibility rather than billability. You're not doing client service. You're supplying the defense exhibits for a trial you're not allowed to attend.

Write the ruler deliberately, because Oliver says the proposal is where satisfaction gets decided. Whatever you promise, imply, or let them assume becomes the yardstick, and the yardstick is set at the moment of maximum sales adrenaline. The discipline is structural, and it's simple to say: the pitch describes the good case, the engagement letter describes the expected case, and delivery aims to beat the letter. Firms that let the pitch write the ruler spend the whole engagement measuring themselves against their own enthusiasm.

Instrument the renewal chain, and don't read silence as satisfaction. Bhattacherjee's continuance question runs quietly, and by Entry 05's logic, its failure mode is rarely a complaint. It's drift: the meeting that gets shorter, the sponsor who stops attending, the response time that stretches. Those are the votes coming in. A quarterly conversation that asks the confirmation question out loud, "is this matching what you expected when you hired us?", is uncomfortable exactly once, and then it's the cheapest retention tool you own.

Time your referral asks to the confirmation peak, because Festinger hands you a gift here: a client who just concluded, publicly, that hiring you was right is motivated to keep proving it, and recruiting a peer is the strongest self-justification available. The referral isn't a favor they do you. Done right, it's a favor you do their story.

Then run the reflection loop on your own shop, and go double-loop at least once a year: win reviews as rigorous as loss reviews, and the assumption-level question nobody enjoys, are we winning for the reasons we think? Use the machines for the measurement, they're better at it. Keep the reckoning human, named, and slightly uncomfortable, because that's the loop your expertise comes out of.

What backfires: celebrating and vanishing. The dinner, the press release, the team reassigned by Friday, every hour of it broadcasting to your new client that the courtship was the product.

The hard call: before you celebrate your next win, calendar the second sale, the ninety-day campaign for the champion who bet on you. The client starts re-deciding tomorrow morning. Be in the room, even when you're not.

Primary Sources

• Leon Festinger, A Theory of Cognitive Dissonance. Stanford University Press, 1957.
• Richard L. Oliver, "A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions." Journal of Marketing Research, vol. 17, 1980, pp. 460–469.
• Anol Bhattacherjee, "Understanding Information Systems Continuance: An Expectation-Confirmation Model." MIS Quarterly, vol. 25, 2001, pp. 351–370.
• Donald A. Schön, The Reflective Practitioner: How Professionals Think in Action. Basic Books, 1983.
• David A. Kolb, Experiential Learning: Experience as the Source of Learning and Development. Prentice Hall, 1984.
• Chris Argyris & Donald A. Schön, Organizational Learning: A Theory of Action Perspective. Addison-Wesley, 1978.