A finding that cannot be recovered cannot support the next step. It cannot guide a design change, strengthen a paper, answer a reader’s question, or show the value of research. This is the problem this lesson addresses: what is lost when findings disappear, and how can simple records preserve their use?
A finding is a piece of evidence from research. It may come from an interview, a usability test, a survey, or a search through a digital collection. The finding might be small. A user cannot locate a button. A search term misses a group of records. A feature helps people finish a task faster.
The finding gains value when someone can use it. That use may happen months later. It may happen in another study, another course, or another product decision. If the record is vague or missing, the finding stops at the moment it was first noticed.
Findings connect research to change
Research work often begins with a question. What do people need? Where do they struggle? Does a new design help them complete a task? A study can answer these questions, but an answer alone does not show impact.
The next link is action. A team uses the finding to change a design, revise a search method, or alter a research plan. Then it checks what happened after the change. This creates a clear chain:
Research showed X.
The team changed Y.
The result shifted by Z.
Without the first part, the chain breaks. People may remember that “the old version was confusing,” but they may not remember which task caused trouble, how many people struggled, or what evidence led to the change.
That gap affects career growth. Researchers often need to explain their work to supervisors, hiring panels, funders, or project partners. A claim such as “the research improved the product” is weak without a record of the findings and the later result.
Academic progress depends on the same link. A student may discover a useful pattern in a database, then lose the search terms and notes. A scholar may identify a source but fail to record its place in the collection. The work then takes longer to repeat. Some of it may never be recovered.
A finding needs a usable record
A usable record does not need to be large. It needs enough detail for another person, or your future self, to understand what happened.
For a database search, that record might include the search terms, filters, date, collection, and reason the result mattered. For a usability study, it might include the task, the user’s result, a rating, and a short explanation.
Simple measures help turn impressions into evidence. Task success asks whether people completed a task. Time on task records how long completion took. Error rate counts mistakes. A satisfaction question records how easy or difficult the task felt.
These measures do not replace explanation. They give the explanation a firm place to begin. A score of 3 out of 5 tells us little by itself. The follow-up question, “Why did you give that rating?” may reveal that the user could not find the search field or did not understand the result labels.
This is also where findings become teachable. A vague note says, “Search was hard.” A useful note says, “The participant found the record, but needed help choosing the date filter.” The second note can guide a change. It can also help another researcher understand the problem.
One small example
Suppose a digital collection has a search page for finding letters. In an early usability session, participants rate the task’s ease of use at 3 out of 5. Their comments point to the same problem: the date filter is hard to see.
The team changes the page. It makes the filter easier to find and gives it a clearer label. A later session uses the same task and the same rating question. The average ease rating rises to 4 out of 5.
The value lies in the record. The finding identified the problem. The change responded to it. The later measure showed a shift. The team can now explain what research contributed.
If the early findings vanish, the change may still look sensible. The proof becomes weaker. No one can tell why the filter changed or compare the old and new experience with care.
The same structure applies to academic database work. A researcher may begin with a broad search and find few useful records. Notes show that one subject term produced better results than another. Later, a collaborator can repeat that path instead of starting over.
Findings support a research career
Career growth depends on visible work. That does not mean every project needs a dramatic result. It means the work needs a clear account.
A record of findings can show how a researcher framed a question, gathered evidence, made a judgment, and checked the result. These are portable skills. They apply to scholarly research, library instruction, digital projects, and user research.
Metrics can help express that work in plain terms. A task success rate may show that more users completed a search. A time measure may show that the task became faster. A satisfaction score may show that the experience felt easier. Each measure answers a different question.
The record also needs limits. A small usability study can reveal why people struggle, but it may not represent every user. A higher rating does not prove that every problem disappeared. A successful database search does not prove that the collection contains every relevant source.
Stating these limits makes the work stronger. It shows judgment. It prevents a small finding from becoming a large claim.
What disappears when findings are lost
When findings disappear, several kinds of progress slow down.
Research becomes harder to repeat. A later study may use different tasks or questions, so the results cannot be compared well.
Design decisions lose their history. People may know what changed but not what evidence led there.
Teaching loses its examples. Students hear that evidence matters, but they cannot see how a small observation shaped a larger decision.
Career records become thin. A researcher may remember doing careful work, yet have little material to show how that work affected an outcome.
The loss can also change the direction of future research. Missing findings create empty spaces. Those spaces may be filled by memory, guesswork, or the loudest opinion in the room.
That is why preservation is part of research, not clerical work after research. The note, score, search string, and explanation carry the work forward.
A practical record can stay small. It can name the question, task, source, finding, change, and later result. It can keep the same measure before and after a change. It can separate what was observed from what was inferred.
This is the kind of work we try to make visible through Beyond Citation. A digital source matters because of what it lets researchers find, compare, and preserve. Its value weakens when the path to a finding cannot be understood later. Our own project is still learning how to describe those paths plainly, including the limits of what a database record can show.
The central lesson is simple. Findings are not the end of research. They are the bridge between evidence and progress. When that bridge is missing, careers, projects, and academic arguments all lose support. When it is recorded well, a small observation can guide a later study, a better search, or a measurable change.
The Source List keeps that same promise in view: one digital source worth knowing, one search tip, and one honest limitation. That balance leaves room for evidence, use, and the gaps that still need work.