Not every number worth trending comes off a sensor. Tank gauges read by hand, samples logged on rounds, and readings from instruments that are not wired to the control system all produce values people care about, and they need a home in the historian alongside the automatically collected tags. A manual data entry tag is that home: a tag whose values are typed in by a person rather than gathered from a device, timestamped and flagged so it is clear the number was keyed in and by whom. This page explains how a manual entry tag is written, how it is timestamped and quality-flagged, and why the audit trail behind it matters as much as the value itself.
Manual data entry tag in one line: A manual data entry tag is a historian tag whose values are entered by a person, through a form or operator screen, rather than collected automatically from a device or controller. Each entry is stored with a timestamp and a quality indicator that marks it as manually entered, and it usually records who entered it and when, forming an audit trail. Such tags let hand-read gauges, rounds data, and off-line instrument readings live in the historian next to live tags, so trends and reports can combine both.
The defining feature of a manual entry tag is its write path. Instead of a collector polling a device and pushing samples, a person opens a form or an operator screen, types a value, and submits it, at which point the value is written to the historian for that tag. The entry usually lets the person specify the timestamp the reading applies to, because a rounds reading taken at ten in the morning might be entered at eleven, and the value should be stored against when it was measured, not when it was typed. That separation between measurement time and entry time is a small but important part of getting manual data right.
Because a human is in the loop, the write path adds checks that automated collection does not need. A good manual entry form validates that the value is within a plausible range, prompts before accepting something that looks wrong, and may require the reading to fit an expected schedule so a missed round is visible. Some tags accept only one entry per period, so an accidental double entry is caught, while others allow correcting a previous entry, which then has to be handled carefully so the correction is recorded rather than silently overwriting history. These guardrails exist because the weakest point of manual data is a typo, and the form is where a typo can be stopped.
Manual entry tags coexist with live tags in the same archive and are queried the same way, which is much of their value. A report can pull an hourly automated flow next to a shift lab result and a daily hand-gauged level, all as tags, without special handling. The historian does not treat the manual tag as second class in storage; it simply records the different origin through quality and audit metadata, so that anyone reading the data later can tell which numbers came from a device and which came from a person.
Every manual entry carries a timestamp for the moment the reading represents, and it carries a quality indicator that marks the value as manually entered rather than collected. That quality flag is not a judgment that the value is bad; it is provenance, telling downstream consumers that a person supplied the number. Analysts and reports can use that flag to treat manual and automated data appropriately, for instance noting that a manually entered figure was not continuously monitored, or filtering to only device-collected values when a calculation needs them.
Beyond value, timestamp, and quality, a manual entry tag typically records an audit trail: who made the entry, when it was actually entered, and if a value was later corrected, what it was before and who changed it. This matters because manual data is exactly the kind that gets questioned later. When a lab result or a hand reading feeds an accounting figure, a compliance record, or a decision, being able to show who entered the value and whether it was ever revised is what makes the number defensible. An audit trail turns a keyed-in number from an unverifiable claim into a traceable record.
Handling corrections without losing history is the subtle part. If someone realizes yesterday's entry was wrong, the right behavior is usually to record a correction that supersedes the original while keeping the original visible in the audit trail, rather than erasing it. That way the history shows both the original and the corrected value with their times and authors, which preserves the integrity of the record. A historian that quietly overwrote the old value would make the data look clean but would destroy the very traceability that manual entry is supposed to provide.
In field operations, manual entry tags are how the parts of the plant that are not fully instrumented still make it into the data. Operators on rounds read gauges, note tank levels, and record the condition of equipment that has no wired sensor, and entering those on a tablet or a workstation puts them in the same historian the SCADA system feeds. That unifies the picture, so a supervisor looking at a trend sees both the continuously monitored variables and the periodic hand readings, and does not have to reconcile a separate paper log with the digital record.
This complements rather than competes with automated collection. A live tag gives continuous coverage of what a sensor can measure, while a manual tag fills the gaps for what only a person on site can observe, or for measurements that require sampling and off-line analysis. The two together give a fuller account than either alone, and because manual entries are clearly flagged, no one mistakes a once-a-shift reading for continuous data. For remote or lightly staffed sites feeding a central monitoring system, manual entry is often the only channel for readings that still require a human visit, and having them flow into the same cloud-visible history keeps a distant team informed.
The discipline manual entry asks for is timeliness and honesty in the record. A reading entered late against the correct measurement time is fine; a reading fabricated to fill a missed round is not, and the audit trail exists partly so that patterns of missing or suspicious entries are visible. Treating manual entry tags as first-class but clearly-marked data, with validated input, honest timestamps, quality that signals provenance, and a correction history that never quietly overwrites, is what lets these hand-supplied numbers be trusted in the same reports and analyses as the automatically collected ones.
A lab data tag focuses on receiving results produced by a laboratory, often delivered from a lab system, whereas a manual data entry tag is about the manual write path itself, a value a person types into a form on an operator screen or tablet. The manual entry concept centers on validated keyed-in input, the separation of measurement time from entry time, and the audit trail of who entered or corrected a value. Lab results can be one source of manual entries, but manual entry also covers rounds readings and hand-gauged levels.
No. The manual entry quality flag is provenance, not a verdict. It tells downstream consumers that a person supplied the value rather than a device collecting it automatically. That distinction lets reports and analyses treat manually entered data appropriately, for example noting it was not continuously monitored or filtering to device-collected values when a calculation requires them, without implying the number is wrong.
The correct approach is to record a correction that supersedes the original while keeping the original visible in the audit trail, along with who changed it and when. This preserves traceability, since the history shows both the original and the corrected value with their times and authors. Silently overwriting the old value would make the data look clean but would destroy the audit trail that manual entry exists to provide.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.