An Admin Team Tests Protected Blocks Against Its Own Backlog

Healthcare IT

Everything below is invented to carry an argument; no real team is being described, and the figures are deliberately small and rough. The argument is about blocking time for deep work in the administrative side of an online health-information service, where the work is not clinical at all: assembling record packs, checking that forms are complete, and reconciling the paperwork that surrounds a consultation. Here is how a small team of four in such a service came to reserve protected blocks, and what they observed when they tested the idea against their own numbers.

The situation: a backlog that nobody could explain

The team handled administrative requests: preparing the document set a user needed to bring to an appointment, checking that uploaded forms matched the checklist, and correcting formatting errors in submitted records. None of it required medical judgement, and all of it required care. A missing page or a mismatched date created a round of back-and-forth that cost far more than the original task.

Over a few months a backlog built up. Requests were answered quickly when they were simple, but the larger packs, the ones needing a careful, page-by-page check, kept slipping. The team lead’s first explanation was staffing. Their second was tooling. Neither survived a closer look, because the team’s logged hours were not short. The hours were simply fragmented.

The evidence: what the team measured first

Rather than act on instinct, the team ran a two-week observation. Each member recorded, on a shared sheet, the start and end of every uninterrupted working stretch on a large pack, along with the reason each stretch ended. They kept the method crude on purpose: pen-and-paper level, no software purchase.

Three patterns emerged from the sheet.

  • Stretches on large packs were short. The typical stretch ended well before the point where a reviewer is fully absorbed in a document set.
  • Most stretch-ending events were small requests from colleagues or users that could have waited.
  • Rework correlated with fragmentation. Packs assembled across many short stretches were more often sent back for a correction than packs completed in one longer sitting. The sample was small, so the team treated this as a hint, not a proof.

The caution mattered. The analyst on the team insisted the write-up say what the data could and could not support: the pattern was consistent with interruptions causing errors, but with two weeks of observations from four people, it could not rule out other explanations such as differing pack complexity.

The hypothesis and the trial design

The hypothesis was specific: if large packs were worked in protected blocks of ninety minutes, they would be completed with fewer returns for correction. To test it fairly, the team set a trial of three weeks with simple rules.

  1. Each member received two protected blocks a week, booked on the shared calendar under the name of the pack being worked.
  2. During a block, a rotating colleague handled the shared inbox and calls, and logged anything that needed the blocked person afterwards.
  3. Urgent items had a written route: a phone call to the covering colleague, who decided whether to break the block.
  4. Packs worked outside blocks continued as before, which gave the team a built-in comparison.

That last rule is what made the trial evidence-led rather than anecdotal. Both methods ran side by side on similar work, and the sheet kept recording.

Week one: a rough start

Opening days went badly, which the team later considered useful. Two blocks were broken by requests that the cover colleague judged urgent but that turned out not to be. A third was lost to an all-hands meeting nobody had cleared in advance. Members reported feeling guilty at being unreachable.

The team adjusted two things. They tightened the definition of urgent to cases where a user’s appointment was within the next day, and they asked the lead to keep the blocks clear of standing meetings. Neither change altered the principle; both made the arrangement fit the real week.

Weeks two and three: the data turns

By the second week most blocks held. Members described a different quality to the work: after the first twenty minutes, there was a stretch of steady progress that they had not felt in months. The sheet recorded longer stretches, and fewer events ended them.

At the end of week three the team compared the packs. Packs completed inside blocks went out with fewer returns than packs worked in scattered stretches during the same weeks. They also took less total time per pack, which surprised the team lead. Again, they labelled the finding preliminary: a small team, a short trial, a possible bias because members chose the most complex packs for their blocks. They noted that last point as a flaw to correct in the next round by assigning packs to blocks at random.

The objection from inside the team

Not everyone was convinced. One member, who handled most of the urgent user contacts, argued that a protected block would simply move interruptions onto the covering colleague and leave the total unchanged. The point was reasonable, so the team turned it into a testable claim. During the trial the cover role rotated, and the sheet logged how many requests each covering person fielded and how many truly needed the blocked colleague afterwards. The count of items passed through to the blocked person was small, because most requests could be answered from the checklist or deferred with a clear reply. The load on the cover was real but manageable, and rotating it meant nobody carried it for long.

That exchange taught the team something about how to run a trial: write down each doubter’s strongest objection before the start, and decide in advance which number would answer it.

What the team decided

The team kept the arrangement and adopted three working rules.

  • Two blocks per person per week stay on the calendar by default.
  • Pack assignment to blocks is random for the next month, to remove selection bias.
  • A short monthly review compares returns and stretch lengths, so the habit keeps earning its place.

They also wrote down what the trial had not shown: it said nothing about quality of care, which is outside the team’s remit, and it did not prove that blocks suit every kind of administrative task. Routine requests continued to be handled in the open flow of the day.

Why the first instinct was wrong

It is worth pausing on why the staffing explanation felt so convincing. When work piles up, adding capacity is the intuitive response, and it is often right. Here it would have been expensive and might have changed nothing, because a larger team working in the same fragmented pattern would simply have produced more fragments. The log redirected the effort toward the real constraint, which was the shape of the day rather than its length.

Lessons any administrative team can borrow

First, measure before you change anything; even a crude log of uninterrupted stretches reveals whether the problem is hours or fragmentation. Second, run the new method alongside the old one so you have something to compare. Third, state limits honestly. A claim that fits the data builds trust; an inflated one gets quietly dropped when the next quarter’s numbers disagree.

Readers who like this kind of practical, small-scale thinking may enjoy the site’s earlier pieces, such as the one on splitting household bills with clear rules and a tracker and the roadmap for remote workers learning DevOps to get hired, both of which turn on the value of a simple routine that you actually stick to.

Planning your own calendar around events

Protected time is not only for desk work. If you plan to attend a workshop, conference or meetup, give it a reserved slot as well, so it does not collide with your focus blocks. Hunting for something worth attending? The MyEventsNow site for Indian cities gathers what is coming up near you in one browsable place, and it keeps a member forum and a blog going alongside.

One practical footnote: the team kept a single page listing what counted as urgent, what could wait until the end of a block and who the covering person was that week. It was pinned beside the calendar and revised after each review, which kept the rules visible without anyone having to remember them.

Closing thought

The team in this story did not discover anything exotic. They noticed that careful work was being sliced into pieces too small to be useful, tested a modest change, and kept it because the numbers they collected supported it. The calendar, in the end, is the cheapest instrument any team owns, and reserving a few hours on it for demanding work is a reasonable experiment for anyone who wants their evidence, and their results, to hold up.