The timing principle in CBT self-monitoring

It's Sunday morning. You had drinks on Friday and Saturday. You open ayodee to log the weekend. You try to remember: Friday , was it three drinks or four? The wine at dinner, the drinks at the bar, and then there was one at the end of the night. Saturday , it was a friend's birthday, so more than usual. Six, maybe? Or seven?

You enter something that feels approximately honest and close the app.

You've kept a record. But you've done significantly less than you could have done , and the reason isn't accuracy, though that's part of it. The reason is timing.

What the research says about timing

The clinical literature on CBT self-monitoring is unusually specific on this point. The recommendation across multiple sources is consistent: self-monitoring should be completed "shortly after , or during , an event" (Korotitsch & Nelson-Gray, 1999). Not at the end of the day. Not the next morning. In the moment, or as close to it as possible.

There are two distinct reasons for this, and they're both important.

The first is accuracy. The second , less intuitively obvious but equally significant , is therapeutic effect.

The accuracy problem with memory

Memory is not a recording device. It's a reconstructive system that rebuilds events from partial information, influenced by mood, expectation, and the narrative that best fits the person's self-image. The implications for self-monitoring are significant.

People systematically underestimate alcohol consumption when recalling it retrospectively. The research on this is consistent: recalled estimates are reliably lower than real-time records across populations and methodologies. The effect is not simply that people forget some drinks , it's that the reconstruction of a drinking occasion involves processes that tend to produce a lower number than the accurate one.

The "unremarkable" drinks are the ones that disappear from memory. The third drink, ordered while the conversation was good. The top-up poured at home before the end of the evening. These aren't blocked from memory deliberately. They're simply not salient enough to survive reconstruction. The memorable moments of an evening , the conversation, the meal, the event itself , are retained. The specific drink count, attached to none of those memories in particular, compresses.

What this means practically: the person who recalls "about four drinks on Saturday" may well have had six. Not because they're being dishonest , because memory doesn't preserve what a record would preserve.

The therapeutic effect of timing

The accuracy argument alone would be sufficient reason to log in the moment. But there's a second, more interesting reason , one that goes to the heart of why self-monitoring is therapeutic rather than merely informational.

The therapeutic mechanism of self-monitoring , the reactivity effect, the way observation changes behaviour , operates at the moment of noticing. The five-second pause at the bar, the act of opening the app while the drink is in front of you, the moment of registering "this is what I'm doing now" , these are where the intervention lives.

Logging the previous night's drinking the following morning doesn't produce that pause. The behaviour has already happened. The sequence of cue, response, and consumption is complete. The logging updates the record, but it doesn't insert the moment of consciousness into the moment of behaviour , which is the therapeutic mechanism.

Consider two people using ayodee:

Person A logs at the bar, in the moment. Each drink is logged as it's ordered. The act of logging introduces a brief awareness , a registration of the drink, the time, the context , into the sequence of ordering and consuming. Across weeks and months of this practice, the repeated insertion of awareness changes the relationship between the situational cues and the automatic behaviour they trigger.

Person B logs the previous day's consumption each morning from memory. The record is approximate. More importantly, the moment of conscious awareness occurs at 8am, in retrospect, while the relevant behaviour occurred at 8pm the previous evening. The logging doesn't interact with the behaviour. It documents it from a distance.

Person A is practising a CBT intervention. Person B is maintaining an approximate record. Both are better than nothing. But the difference in therapeutic value is substantial.

The accessibility of in-the-moment logging

The historical barrier to in-the-moment self-monitoring was practical. Paper records require something to write on and something to write with, in a context where neither is typically present. The printed NHS worksheet doesn't travel to the bar. The counsellor's intake form doesn't accompany you to Friday dinner.

A phone-based diary removes this barrier. The phone is present in the situations where the behaviour occurs , at the bar, at dinner, in the kitchen, at a friend's house. The app is three taps away. The log entry takes twenty seconds.

This is not a trivial design consideration. The clinical literature notes that "the method and format of self-monitoring can be adapted to suit each client's preferences" and that if a format is impractical , if it requires resources or tools that aren't available in the relevant context , accurate recording is compromised. A tool that lives in your pocket and takes twenty seconds to use is one that can reasonably be used at the moment the behaviour occurs.

What to do with the "I forgot to log" problem

The timing principle doesn't require perfect real-time recording to produce value. What it suggests is a hierarchy of approaches, ordered by therapeutic effectiveness:

Log in the moment, at the point of consumption. This is the gold standard and the practice that produces the strongest therapeutic effect.

Log shortly after , within an hour or two of the occasion. This preserves most of the accuracy benefit and some of the timing benefit. The memory is fresh, the session is recent, the context is still present.

Log the same day, before bed. Accuracy is reduced relative to real-time but substantially better than the following morning. The therapeutic effect of the pause is absent, but the record is useful.

Log retrospectively, from memory. Better than nothing. Produces approximate data that's more useful than no data, but significantly less useful than the alternatives above.

The goal is not perfection. The goal is shifting practice toward the left end of this spectrum , toward recording in the moment, in the relevant context, as often as possible. Each shift in that direction produces more accurate data and more therapeutic effect.

ayodee is built for in-the-moment logging , it's on your phone, it takes under a minute, and it works in the places where the behaviour happens. That's not convenience. That's where the mechanism lives.

References Korotitsch, W.J., & Nelson-Gray, R.O. (1999). An overview of self-monitoring research in assessment and treatment. Psychological Assessment, 11(4), 415.

Kennerley, H., Kirk, J., & Westbrook, D. (2017). An Introduction to Cognitive Behaviour Therapy: Skills & Applications. 3rd ed. Sage.

Bornstein, P.H., Hamilton, S.B. & Bornstein, M.T. (1986). Self-monitoring procedures. In A.R. Ciminero et al. (Eds.), Handbook of behavioral assessment (2nd ed.). Wiley.