7 Points to Check Before Conversion: Why CAD and GIS Coordinate Systems Don't Match

When attempting to link CAD and GIS, it is not uncommon to encounter problems such as drawings looking fine but not overlapping the background map, positions shifting relative to existing data, appearing rotated, or being placed in a location far away. When these malfunctions occur, it’s easy to assume that the conversion settings have failed, but in many cases the real cause is a lack of checks before the conversion process itself. In other words, it’s not that the coordinate system doesn’t match, but that the prerequisites about what to align and how to align it have not been established before starting the work.

Especially in practical work, the mindset of CAD, which centers on design drawings, and the mindset of GIS, which centers on positional information, tend to become mixed. CAD emphasizes consistency within the drawing, whereas GIS emphasizes consistency with real-world space. As a result, even when looking at the same lines or points, the meaning of the coordinates can change depending on what assumptions each person in charge is making. Precisely for that reason, to understand why coordinate systems do not match between CAD and GIS, it is important to grasp the points that should be checked before conversion, rather than the procedural steps of the conversion itself.

In this article, I organize and explain seven causes you should review when coordinate systems don’t align between CAD and GIS, presenting them as seven checkpoints to verify before conversion. The intended readers are practitioners who work across drawings and geographic information. Rather than packing in difficult theory, the focus is on where to look to reduce rework and how to think about ensuring safe interoperability.

Table of Contents

• Why coordinate system mismatches frequently occur between CAD and GIS

• Checkpoint 1: Is the origin a local coordinate or a geographic coordinate?

• Checkpoint 2: Are the assumed units consistent?

• Checkpoint 3: Are planar coordinates being confused with latitude/longitude?

• Checkpoint 4: Are the geodetic datum and projection specified clearly?

• Checkpoint 5: Are the drawing rotation and the north orientation consistent?

• Checkpoint 6: Are the height reference and vertical datum consistent?

• Checkpoint 7: Have known control points and their correspondences been verified?

• Practical sequence of checks to perform before conversion

• Summary

Reasons why coordinate systems frequently don't match between CAD and GIS

CAD and GIS both deal with points, lines, and areas, so they may appear to manage location in the same way. However, their purposes are slightly different. CAD excels at accurately drawing structures and plans and at refining dimensions and shapes. GIS, on the other hand, excels at managing what is located where in real space and at overlaying that information with other spatial data. This difference in purpose is directly reflected in their differing approaches to coordinate systems.

In CAD, there are many situations where a drawing is valid as long as the positional relationships within the drawing are correct. For example, if you take the center of a structure as the origin and shapes are accurately drawn by their distances and angles from that point, the drawing can adequately serve its purpose. What is emphasized here is the relative relationships among the shapes. Where they are located on the actual Earth can be supplemented later with other information.

In GIS, it is necessary to manage where data are located in the real world according to a common frame of reference shared with other information. To overlay data from different sources—topographic maps, roads, rivers, administrative boundaries, parcel numbers, aerial photographs, asset registers, and so on—onto the same location, the positional reference must be unambiguous. Therefore, in GIS it is important that prerequisites such as coordinate systems, geodetic datums, map projections, units, and orientation are all consistent.

The problem is that, in practical work, these differences are not neatly separated. Some CAD drawings are based on surveying results, and some geographic data have been processed as design drawings. Therefore, it's dangerous to judge solely by appearance that something is one way because it’s CAD or another way because it’s GIS. What you need to do is confirm what the data is using as a reference to express positions. If you feel the coordinate systems don't match, it's important to identify the assumptions of the data itself one by one before looking at the conversion tool's settings screen.

Checkpoint 1: Is the origin in local coordinates or geographic coordinates?

The first thing to check is where the origin or reference position of the data is located. If you begin conversions without knowing this, all subsequent work becomes guesswork. This is the most fundamental—and also the most frequent—cause of coordinate mismatches between CAD and GIS.

In CAD, an arbitrary point is sometimes set as the origin to prioritize ease of handling the drawing. The origin may be a point convenient to the drafter, such as the center of a structure, a corner of the site, the intersection of street centerlines, or the position of a reference stake. In this case, if distances and angles within the drawing are accurate, the drawing is sufficiently usable. However, as it stands, it is not clear where it is located in the real world.

On the other hand, GIS generally assumes coordinates that are tied to real-world space. Whether latitude/longitude or planar coordinates, they are used to indicate positions on the real Earth. In other words, GIS coordinates need to be on the same footing as other map data. If the origin of a CAD drawing remains local, importing it into a GIS as-is will not necessarily cause it to align to the correct position.

A common tendency in practice is to assume “it’s a survey drawing, so it should be fine,” or “it’s a design drawing, so the positions must be correct.” However, even when the survey results are the basis, they may have been rearranged into a local coordinate system for design or drafting convenience. Conversely, CAD data may have been drawn to match the public coordinate system from the outset. What matters is not the appearance or the file format, but confirming which origin the drawing is based on.

A useful way to verify is to check whether there are known points. Confirm whether the reference points on the drawing correspond to known points or coordinate values on site, and whether you can explain which point on the drawing corresponds to which point in real space. If you cannot explain this, it is not yet suitable for use as geographic coordinates. Converting it while the meaning of the origin is ambiguous may only place it near the correct location by chance, and cannot be considered truly correct.

If you skip this check, you will likely end up forcing a fit later by repeatedly translating and rotating. However, data fitted in such a makeshift way has low reproducibility and cannot be handed over to other team members. Determining whether the origin is local or in geographic coordinates is the first branching point before conversion. If this remains ambiguous, no matter how carefully you perform the conversion, it will not lead to stable data exchange.

Check Point 2: Are the assumptions about units consistent?

The next thing to check is the units. Differences in coordinate systems may sound difficult, but many of the misalignments that occur in practice are due to differing understandings of units. And because unit issues appear elementary, they are even more likely to be overlooked.

In CAD, drawing units such as millimeters (mm / in), centimeters (cm / in), and meters (m / ft) are used depending on the project and operational rules. If dimensions are correctly represented on the drawing, the purpose of drafting is fulfilled, so which units the internal numerical values are in may only be shared as an organizational rule. Therefore, when handed over to another person or another system, the interpretation of the units can become inconsistent.

In GIS, units are directly tied to the meaning of positions: degrees for latitude/longitude, and usually meters for projected coordinates. If units become inconsistent on the GIS side, data will no longer align with other map layers. In other words, units are not merely a drafting convenience but a fundamental requirement for spatial consistency.

For example, if a value of 1000 in CAD is meant to represent 1 m (3.3 ft) but the GIS interprets it as 1000 m (3280.8 ft), the overall size and positioning of the geometry will be greatly distorted. Conversely, if meter (m) coordinates originating from GIS are handled on the CAD side as if they were in millimeters (mm), the numbers will appear extremely large and become difficult to work with.

When checking, you need to look at both the coordinate values and the geometric dimensions. Examine known road widths, structure lengths, site dimensions, etc., and verify whether those numbers correspond to real-world sizes. If one side of a drawing is 5000, you can, by practical judgment, determine whether that is natural as 5 m (16.4 ft) or implausible as 5000 m (16404.2 ft). This kind of intuitive check helps to detect unit mismatches early.

Also, attention is required for cases where the units of coordinates and the units of drawing dimensions do not necessarily match. It is possible, for example, that placement positions are based on meters (m / ft) while the shape dimensions assume millimeters (mm / in). In such cases, even if the drawing itself is valid, sufficient explanation is needed when coordinating with external parties. Units are not necessarily uniform; it is important to be mindful to check them for each item, such as coordinates, dimensions, distances, and heights.

Discrepancies in units are problems that are easy to fix once discovered. However, if you proceed with conversions while overlooking them, they can later appear as inconsistencies in scale or position and make the root cause hard to identify. Especially when coordinate systems between CAD and GIS do not align, suspect the units first. Simply adopting this basic stance can greatly reduce unnecessary rework.

Checkpoint 3 Are you confusing planar coordinates with latitude and longitude

The third point to check is the confusion between planar coordinates and latitude/longitude. This is basic on the GIS side, but for staff who mainly work with CAD it is easy to regard them ambiguously as just another type of location information. However, if these are mixed up, the data may not only fail to overlap but may appear in a completely different place.

Latitude and longitude are a way of expressing positions on the Earth as angles. By contrast, planar coordinates are numerical values that represent the Earth's surface on a plane through a projection, making distances and positions easier to handle. Both represent locations, but the nature of the numbers and their units are different. Treating latitude and longitude values as if they were planar coordinates will cause inconsistencies, and the reverse is also true.

In CAD, shapes are typically handled as XY coordinates on a plane. Therefore, if it is not made clear what the numbers are based on, they all appear to be the same kind of coordinates. In GIS, some data are handled as latitude and longitude, while other data are converted and used as planar coordinates, such as the plane rectangular coordinate system. If these assumptions are not aligned, large discrepancies can occur the moment the data are imported.

A common situation in practice is that, in conversations between those responsible, the word "coordinates" is used alone and the type of coordinates is omitted. One person may be referring to latitude and longitude while another assumes planar coordinates, yet the transfer proceeds as is. As a result, positions don't match, and they end up repeatedly trying conversion settings without understanding the cause.

As a method of verification, looking at the format of the numbers is effective. By checking whether they are angle-like values that include decimals or large numbers to be treated as distances, you can get a rough idea. However, you cannot determine it from appearance alone, so you need to confirm which map or known points the coordinates correspond to. For example, it is important to overlay them on existing topographic maps or base maps to see whether the nature of the coordinates can be explained.

The reason this check is important is that the difference between planar coordinates and latitude/longitude is not merely a difference in display format. Depending on the task, it directly affects distance calculations, area calculations, positioning, and overlay accuracy. To safely integrate CAD and GIS, you need to concretize what is meant by the term "coordinates" and clarify whether you are dealing with angular coordinates or planar distance coordinates.

Checkpoint 4: Are the geodetic datum and projection specified clearly?

The fourth point is the specification of the geodetic datum and projection. This may seem a bit technical, but it is unavoidable if you want to handle GIS locations correctly. When the coordinate systems in CAD and GIS don't match, the numbers themselves may look plausible yet not align. In such cases, this specification may have been passed along ambiguously.

A geodetic datum is the concept of which reference is used to represent positions on the Earth. A projection is the method of representing those positions on a plane. In GIS, these two together underpin the meaning of location. In other words, XY values alone are insufficient; unless you know which datum and which projection those values are based on, you cannot safely overlay them with other data.

In CAD, because the primary purpose is producing detailed drawings, this information may exist outside the file. Even if it has been shared among the drafter and other stakeholders, someone who only looks at the file may not be aware of it. For example, you might intend the drawing to be based on a public coordinate system, but without an explanation of the system number or reference, the recipient can only guess. If their guess happens to be correct, that's fine; if not, the position will be offset.

In practice, you may assume you are using the plane rectangular coordinate system only to find that a different system was actually used, or that the data were prepared to a different reference than the background data. In such cases, it is dangerous to be reassured by the apparent closeness of the numbers. Even if the offset seems small, the error becomes more noticeable as the area increases and can cause major problems in later stages.

When checking, it is important not only to look at the coordinate values themselves but also to review the setup documents, handover conditions, drawing notes, and related survey materials to confirm which reference standard is being used. It is also important that the information is preserved in a reproducible form, not merely conveyed orally. If another person later cannot reproduce the same transformation, the practical coordination is insufficient.

Checking the geodetic datum and projection is not about memorizing difficult theory. What matters is making the meaning of positions reproducible by others in the same way. In the integration of CAD and GIS, it is the clear specification of these preconditions—rather than the transformation itself—that determines quality. If you proceed while the specifications remain ambiguous, the transformation results can easily appear to match only by coincidence, so it is important to address this at the outset.

Checkpoint 5: Are the drawing rotation and the north direction aligned?

The fifth issue is the rotation of the drawing and the direction of north. In cases where people report that CAD and GIS coordinate systems don't match, the problem can actually be the drawing's orientation rather than a coordinate transformation. If positions are generally close but offset slightly at an angle, or the offset grows larger the farther away you go, you should suspect this.

In CAD, drawings are sometimes placed at arbitrary angles to make them easier to read. For reasons such as wanting to display the main axis of a road or structure horizontally, to make the drawing fit better on the sheet, or to make drafting and dimensioning easier, the entire drawing may be rotated. This is a natural way of representing drawings and is common in practice.

In GIS, it is fundamentally assumed that orientations are consistent with those of the real world. Because background maps, terrain, administrative data, and other facility information are overlaid, north serves as the reference for spatial alignment. Even if you rotate the view on screen, the positional relationships in the data itself must follow geographic orientation. This is the difference from CAD, which prioritizes the readability of drawings.

The problem occurs in cases where a CAD drawing has been rotated but this goes unnoticed and it is overlaid onto a GIS as-is. In such cases, even if you align a single point, the whole will not match. It may appear to overlap near the center, but it will be significantly misaligned at the edges. Moreover, when the extent is small the discrepancy is less noticeable, so it is easy to overlook.

When verifying, it is easier to judge if you compare with features whose orientation is known in the real world, such as road centerlines, parcel boundaries, revetment lines, and building edges. Also, just because the drawing frame or the orientation of text is neatly arranged does not mean it corresponds to true north. The more readable a drawing is, the more likely it has actually been rotated arbitrarily.

Furthermore, when an object is rotated, it is also important which point it is rotating about. Whether you rotate around the origin or around a reference point will change the final placement. Therefore, you need to check not only the north orientation but also the concept of the reference point and the center of rotation.

When you say that the coordinate systems don't match, people tend to focus only on numerical conversions. However, in practice, significant inconsistencies can arise simply from different assumptions about orientation. Confirming a drawing's rotation and the direction of north before converting is an important step to avoid being misled by apparent visual alignment.

Checkpoint 6: Are the height reference points aligned?

The sixth issue is the vertical reference. When people talk about integrating CAD and GIS, attention tends to focus only on horizontal positions, but in practice mismatches in elevation are also a major problem. Even if plan views appear to overlap, if the vertical reference differs, it causes problems for cross-sections, earthwork quantities, as-built geometry, construction planning, and maintenance and management.

In CAD, the way height is represented varies depending on the type of drawing—plan views, elevations, sections, three-dimensional models, and so on. Height may be stored as an attribute of a geometry, or it may be recorded only as an annotation. In other words, simply checking whether CAD data contains Z values is not sufficient to determine the meaning or reliability of the height information.

In GIS, height is handled in various forms such as ground elevation, structure height, elevation, and relative height. Furthermore, the horizontal position reference and the vertical reference are sometimes managed separately. Therefore, it is risky to assume that Z is also correct just because the XY coordinate system aligns.

A common issue in practice is being satisfied with checking only the planimetric position. People feel reassured when points align with the background map, but if they later find that the elevations don't match, the burden of readjustment becomes significant. In particular, in fields where elevation differences directly affect the work—such as land development, roads, slopes, rivers, and areas around structures—you must not postpone checking heights.

When checking, it is necessary to clarify what that elevation represents. Determine whether it is the ground surface, the top surface, the difference from the reference plane, the design elevation, or the measured elevation. Also, confirm which reference plane that elevation is based on. Even if the numbers match, the meaning changes if the reference is different.

In addition, when transferring from CAD to GIS, it is necessary to consider how to handle height representations on drawings as spatial data. Conversely, when transferring from GIS to CAD, the challenge is how to reinterpret height information used for analysis as drawing or construction information. Heights are items whose relationship to business objectives is stronger than that of planar positions.

Downplaying height as a cause of coordinate system mismatch can lead to the largest rework in later stages. The idea that it is enough for only the horizontal plane to match is inadequate for the data integration that supports on-site operations. Before performing any transformation, it is necessary to verify alignment, including the height reference.

Checkpoint 7 Have you verified the known points and their correspondences?

The seventh is confirming known points and their correspondences. Even if you understand all the points discussed so far, if you omit verification of known points during actual data exchange, you cannot guarantee the validity of the results. This check is the final deciding factor in determining the cause of mismatched coordinate systems between CAD and GIS.

A known point is a point that can be recognized as the same both on drawings and in physical space. Reference points such as control points, boundary points, corners of structures, center points, and distinctive points of existing equipment—points whose positions are clear and can be rechecked—fall into this category. With such points, you can confirm positional consistency as a concrete correspondence rather than merely judging whether they visually overlap.

In practice, people sometimes assume it's okay because the overall appearance is similar. However, aligning a single point does not necessarily mean the entire thing is correct. Whether a mere translation will align it, whether rotation is needed, whether there are scale differences, or whether there are local distortions cannot be determined without verifying multiple known points. If there are too few known points, there's a risk it may appear to match merely by chance.

The importance of verifying known points also lies in reproducibility. To ensure the same alignment can be achieved regardless of who performs the work, it must be possible to explain which points were used and how the adjustments were verified. If this cannot be done, alignment results will change each time the person in charge changes, causing confusion with every update or modification. Because data integration is not a one-time task, it is important to document it as a reproducible procedure.

When checking, it's important to use multiple points that are as far apart as possible. Relying only on closely spaced points makes it difficult to detect rotation or distortion. By verifying with points that have different positional relationships—such as opposite edges of the target area, or the center and the periphery—you can more easily judge the overall consistency.

Furthermore, a known point that only has coordinate values is insufficient. It is necessary to be clear which point it corresponds to on the drawing and which object it corresponds to on site. If the correspondence remains ambiguous, simply matching the numerical values does not guarantee accuracy. Only by confirming the meaning of the point does it function as a known point.

The reason the known point was placed last among the pre-conversion checkpoints is that it serves to tie all the other checks to reality. After confirming the origin, units, coordinate type, geodetic datum, rotation, and height, only if they are consistent with the known point is it worth proceeding to the transformation. Conversely, if this is ambiguous, there will remain uncertainty no matter which settings are chosen.

Order of checks to be carried out in practice before conversion

We have reviewed seven checkpoints so far, but in practice what matters is not only what to check but the order in which to check them. If the order is wrong, the underlying assumptions can collapse later and you will have to redo the work. Pre-conversion checks should be carried out not as configuration tasks but as a process to organize the preconditions.

The first thing to do is confirm the origin and intended use of the data. Depending on whether it is a design drawing, something based on survey results, or spatial data for management, the way coordinates are considered will differ. Understanding this reveals the potential for local coordinates, the potential for public coordinates, and how to think about the required accuracy.

Next, verify the origin and units. Check whether the origin is arbitrary or tied to a known point, what the units are, and whether there are any discrepancies between coordinate values and the geometry’s dimensions. If ambiguity remains at this stage, all subsequent conversions will be provisional. Proceeding with provisional assumptions will prevent others later in the workflow from reproducing the results.

Next, confirm whether the coordinates are plane coordinates or latitude/longitude, and whether the geodetic datum and projection are specified clearly. Here, organize the coordinate type and the positional reference frame, and, if necessary, consult related documents or drawing notes to supplement them. Do not be satisfied with the generic term “coordinates”; it is important to know exactly which reference and which format are being used.

Next, verify the drawing's rotation, the direction of north, and the height reference. Don't rely solely on aligning the plan position; also check the orientation and the assumptions about Z. Especially for projects that cover a wide area or where elevation differences are important, do not skip this verification.

Finally, verify overall consistency using multiple known points. Rather than matching just a single point, check whether the results are valid across the entire range. Only after confirming this does it make sense to decide on the transformation settings. Conversely, before you start trial-and-error on the settings screen, it’s better to solidify the assumptions in this order, as that will make the work faster and more accurate.

In practical work, when you’re in a hurry it’s tempting to start with the conversion operation. However, the more rushed you are, the more likely that omitting verification of assumptions will lead to having to redo work later. To reduce the causes of mismatches between CAD and GIS coordinate systems, the most effective measure is to make pre-conversion checks a standard procedure. Instead of relying on experience to make adjustments on the spot, creating a workflow that enables anyone to perform the same checks leads to continuous quality improvement.

Summary

The causes of mismatched coordinate systems between CAD and GIS are not limited to mistakes in transformation settings. Rather, in many cases the root cause is insufficient checking before transformation. The seven points organized here—namely whether the origin is local or geographic, whether units are consistent, whether planar coordinates and latitude/longitude are being confused, whether the geodetic datum and projection are clearly defined, whether the drawing rotation and the direction of north match, whether the vertical reference for elevations is aligned, and whether known points and their correspondences have been verified—by addressing just these seven, many problems with mismatched coordinate systems can be more easily prevented at the pre-transformation stage.

In practice, it’s not a simple matter of “CAD is right” or “GIS is right.” CAD excels at consistency as a drawing, while GIS excels at consistency with real-world space. What matters is not forcing one to conform to the other, but clarifying the underlying assumptions of both and deciding where and how to connect them. To do that, you need not only knowledge of conversion operations but also the perspective to interpret the meaning of the data.

In particular, in practical work that spans design drawings and on-site locations, merely aligning numbers at the desk is not sufficient. Even if things appear correct on the drawings, they are meaningless if they cannot be used in the field, and even if elements overlap on a map, work will not progress if the drawings lack sufficient meaning or contextualization. That is why it is important to carry out careful pre-conversion checks and to connect the site, the drawings, and the geographic information.

In that sense, rather than leaving coordinate checks to desk work alone, having an environment where high-precision positions can be immediately verified on site greatly improves the reliability of decisions. For example, by utilizing LRTK (iPhone-mounted GNSS high-precision positioning device), it becomes easier to confirm the positional relationship between reference points on drawings and their actual positions in the field, and it helps ensure CAD and GIS integration doesn't stop at conversion work but is carried into practical workflows by verifying on site. For those responsible for reducing confusion caused by differences in coordinate systems and improving the accuracy of data integration, adopting such on-site verification methods is a highly practical improvement.