# The Real Cost of Time Leakage for Solo Professionals and Virtual Consultants Airtym · March 12, 2026 · Updated August 24, 2026 > The countable leaks scale with how many clients you have. The expensive one scales with how long you keep them, and it lands on your best relationships. ## What does time leakage actually cost a solo consultant? {#what-does-time-leakage-actually-cost-a-solo-consultant} **The countable leaks — invoicing, rescheduling, chasing payment — are flat. They scale with how many clients you have, not how long you have had them. The expensive leak is the reconstruction before each recurring session: working out where you left off. A first session has nothing to rebuild. A ninth has eight sessions' worth.** Most advice about time leakage counts tasks. Invoicing takes twenty minutes, rescheduling takes ten, chasing a late payment takes three emails. Those numbers are real, and they are the same whether a client is in their first month or their third year. There is a second leak that does not stay the same. Before a recurring call you re-read something — an old email thread, a folder of notes, a calendar entry with a two-word title — to work out what was agreed and what you promised. That reading gets longer with every session you add. It is worth measuring separately, because it grows precisely where your revenue is concentrated. **[Key takeaway]** Admin scales with client count. Reconstruction scales with client tenure. Only one of them gets worse as the practice gets healthier. ## Why does re-reading cost more than it looks like it should? {#why-does-re-reading-cost-more-than-it-looks-like-it-should} Because resuming a suspended task is not the same as continuing one. Erik Altmann and J. Gregory Trafton timed the gap directly, measuring how long people took to act after an interruption ended. The resumption lag ran roughly double the interval between uninterrupted actions, 3.8 seconds against 1.9 ([Altmann and Trafton, Proceedings of the 26th Annual Conference of the Cognitive Science Society](https://www.interruptions.net/literature/Altmann-CogSci04.pdf)). Those are seconds, in a laboratory task, and they do not convert into billable hours. Anyone quoting a tidy number of minutes lost per interruption is extrapolating past what the research supports. What does transfer is the mechanism: picking a task back up costs more than never having put it down, and the cost comes from rebuilding the goal you were holding. The same study found what reduces it. Cues available before an interruption improved performance after it, which the authors read as people deploying "preparatory perceptual and memory processes" to mitigate the disruption ([Altmann and Trafton](https://www.interruptions.net/literature/Altmann-CogSci04.pdf)). Something outside your head carries part of the goal so your memory does not have to. A three-week gap between client sessions is an interruption. Whatever record exists when you sit back down is the cue. ## Why does the cost grow with tenure rather than client count? {#why-does-the-cost-grow-with-tenure-rather-than-client-count} Because what has to be reconstructed accumulates. Ten first-time clients give you ten reconstructions of nothing. One client across ten sessions gives you a history that has to be re-entered before each call, and it is longer every time. That inverts the usual framing. Time leakage is normally described as a volume problem, fixed by handling more clients more efficiently. Reconstruction is a depth problem. Taking on more clients does not make it worse. Keeping them does. ## Why does it land hardest on your best clients? {#why-does-it-land-hardest-on-your-best-clients} Because repeat work is where consulting revenue is meant to sit. David A. Fields puts the healthy band for repeat business — current clients engaging you again after previous work — at 40 to 80 percent of revenue, and treats anything under 40 percent as a problem to fix ([David A. Fields Consulting Group](https://davidafields.com/whats-the-right-amount-of-repeat-consulting-business/)). So a practice built the way it is supposed to be built draws most of its revenue from exactly the relationships where reconstruction costs the most. The leak is a by-product of the thing going right. **[Key takeaways]** If 40 to 80 percent of revenue is repeat work, then 40 to 80 percent of your sessions open with re-reading. ## Where does that record actually live? {#where-does-that-record-actually-live} Usually in your head, with fragments spread across whichever tool handled each event. The notetakers have improved on this. Fathom supports folders for grouping calls, though only on the Team Edition, added one call at a time, with bulk adding not supported ([Fathom Help Center](https://help.fathom.video/en/articles/295808)). Granola describes organising every meeting for a specific account into one folder ([Granola](https://www.granola.ai/blog/best-ai-notetaker-customer-success-teams-crm-integration)). In both cases the filing is yours to do, and it covers the call only. Booking, payment and the video room each keep a separate record of the same relationship, indexed by date rather than by person. When the connective tissue lives in one person's memory, it travels with that person. Umit Gurun, Noah Stoffman and Scott Yonker found that roughly 40 percent of client assets follow a financial adviser who moves to a different firm ([Journal of Financial Economics 141(3)](https://stoffprof.com/GSY21)). The relationship was the asset, and it was portable because the knowledge of it was. [See how a per-client record works →](/signup) ## What actually reduces it? {#what-actually-reduces-it} The research points at cues rather than discipline. You are not going to remember more. The aim is to need less from memory. - Keep one record per client rather than one per event. The person is what recurs, so the person is the right unit to file under. - Write the next step at the end of a session instead of reconstructing it at the start of the next one. The cue costs almost nothing to leave and a great deal to rebuild. - Put the summary where the next session begins. A record you have to go and find is a second interruption. - Audit the leak on your longest relationships, not your newest. That is where it is largest, and where it is least visible. None of this needs a new category of software. It needs the record organised around the thing that repeats. ## Frequently Asked Questions {#frequently-asked-questions} ### Is time leakage the same as poor time management? {#is-time-leakage-the-same-as-poor-time-management} Not for the part that matters here. Scheduling discipline addresses how you allocate hours. Reconstruction is a memory cost that arrives before the work starts, and it persists no matter how well the calendar is arranged. The research points to external cues rather than effort as the thing that reduces it ([Altmann and Trafton](https://www.interruptions.net/literature/Altmann-CogSci04.pdf)). ### How much time does an interruption really cost? {#how-much-time-does-an-interruption-really-cost} Less than the widely repeated figures suggest, and measured in a way that does not translate cleanly to consulting work. Altmann and Trafton recorded a resumption lag of 3.8 seconds against a 1.9-second baseline in a controlled task ([Altmann and Trafton](https://www.interruptions.net/literature/Altmann-CogSci04.pdf)). The useful finding is the direction and the remedy, not a number to multiply by your hourly rate. ### Do AI notetakers solve the reconstruction problem? {#do-ai-notetakers-solve-the-reconstruction-problem} Partly, and only for the call. Per-client grouping exists — Fathom's folders are Team Edition, added one call at a time, with no bulk option ([Fathom Help Center](https://help.fathom.video/en/articles/295808)) — but you do the filing, and the booking and payment records stay outside it. The summary improves; the assembly does not. ### Why does this matter more for repeat clients than new ones? {#why-does-this-matter-more-for-repeat-clients-than-new-ones} Because a first session has no history to rebuild and a tenth has nine. With repeat business sitting at 40 to 80 percent of a healthy consulting practice's revenue ([David A. Fields Consulting Group](https://davidafields.com/whats-the-right-amount-of-repeat-consulting-business/)), most of your billable work happens on the expensive side of that line. [Airtym](/) keeps booking, the video room, payment and the session summary on one running record per client, so the next conversation opens with the last one already in front of you. ## Related Articles {#related-articles} - [Meeting Notes Apps for Client Calls: Two Questions Before You Pick One](/blog/meeting-notes-app-for-client-calls) — per-client grouping exists in every major notetaker; what decides it is who does the filing, and what none of them hold. - [The Tool Tax Nobody Counts: Your Stack Forgets Your Clients](/blog/tool-tax-your-stack-forgets-your-clients) — what the assembled stack costs per month, and the larger cost of every tool in it filing by event rather than by client. - [The Client You Lose Isn't the One Who No-Shows](/blog/client-you-lose-isnt-the-one-who-no-shows) — why the repeat client who stops booking costs more than the one who misses a session, and produces no signal at all.