The short version
- Atlassian's State of Teams 2025 puts 25% of the working week into searching for information and answers.
- The word "answers" is doing a lot of work in that sentence. A large share of the searching ends in asking a colleague, and then in waiting.
- Better search does not shorten the wait, because the thing being looked for was never written down.
- Measure it on your own team by logging searches that ended in a question to a person. That is the number you can act on.
There is a familiar statistic about how much of the working week disappears into looking for things. It gets quoted in every pitch for a search tool, a wiki, or an internal portal, and the implied solution is always the same: index more, search better.
25%
of the working week goes to searching for information and answers
Atlassian, State of Teams 2025
The number, and what it contains
Read the phrasing carefully: information and answers. Those are two different things and they fail in two different ways.
Information is a document, a spec, a policy, a file. When that is hard to find, the fault is organisation, and better search genuinely helps.
An answer is something someone knows. Where a piece of work stands. What was agreed with the client. Why a thing is set up that way. No amount of indexing surfaces those, because they were never written anywhere to be indexed.
What people are actually looking for
Sit with the searches for a week and the pattern is consistent. People start in the document store, fail, search their messages, half find something from four months ago, and then ask a person. The searching was a preamble to asking.
What they are looking for falls into three groups.
- Current state. Where something is right now. Changes weekly, so it is never in a document.
- Reasoning. Why a choice was made. Lived in a conversation, and the conclusion was recorded without the reasoning.
- Commitments. What was promised, to whom, when. Made on a call, written down inconsistently.
All three live in people. This is why teams keep buying search tools and keep feeling that nothing improved: the tool solved the smaller half of the problem.
The wait is the real cost
When the search ends in a question, the clock does not stop at the end of the search. It stops when the person answers.
In one office that is a few minutes. Across time zones it is until tomorrow. On a colleague's holiday it is next week. So the searching time is the visible part and the waiting time is the larger part, and the waiting time does not show up in any statistic about search.
Worse, the waiting is not idle in a clean way. People switch to something else, half finish it, come back, re read what they were doing. The arithmetic is in the cost of context switching and the coordination tax.
Measure it on your own team
General statistics start conversations and do not justify changes. Your own number does.
For one week, ask people to log only the searches that ended in asking a person. Three fields: what you were looking for, who you asked, how long until you could carry on. Nothing else.
You will get two useful outputs. The total waiting time, which is usually larger than anyone expected, and the concentration: a small number of names on the receiving end of most of it. That second output is the same finding as the key person audit, arrived at from the other side.
What actually reduces it
Three things, in order of effect.
Answer in public. The cheapest habit available. Answering in a channel rather than a direct message makes the next identical question findable. It costs nothing and it compounds.
Record reasoning, not just outcomes. One line about why the alternative was rejected prevents a whole category of repeat questions. See the decision log template.
Make current state answerable without the person. The largest group and the one no document store handles. This is what StandIn is built for. Each day, a person spends about ninety seconds on a brief: what moved, what is open, what is blocked, what is next, mostly drafted from the work that already happened. When a colleague needs to know where something stands, they ask and get an answer from that brief, in the owner's words, with a source under it. It never guesses, and when the answer is not in the record it says so and names who to ask. The search that used to end in a question and a wait ends in an answer. See how it works.
Common Questions
Would enterprise search fix this?
It fixes the documents half, which is real. It cannot answer a question about something that was never written down, and it has no way of telling you that it does not know. See enterprise search alternatives.
Is a better wiki the answer?
For stable information, yes. For current state, no, because nobody updates a wiki weekly and a stale wiki is worse than an empty one. See why knowledge bases die in month four.
How do we get people to write more down?
Reduce what has to be written rather than exhorting. The sustainable version is a short daily record attached to the work, not a documentation initiative. See why write it down cultures stop.
Does this get worse as we grow?
It gets worse faster than headcount, because the number of pairs of people who might need each other grows much faster than the number of people. That is why it feels sudden somewhere around thirty.
When you're off, your StandIn is on.
It answers your teammates' questions from work you've already done, in your words, with a source under every answer.