If your child's spreadsheet suddenly shows a circular-reference warning, adding another formula or hiding the warning will not explain what went wrong. Start by tracing where the result gets its inputs. A circular reference becomes manageable when the learner can point to the calculation that eventually asks itself for its own answer.
A spreadsheet circular reference occurs when a formula depends on its own cell, directly or through a chain of other cells. This guide teaches how to recognise the loop, trace the dependency, repair an accidental total, separate inputs from outputs, and distinguish a deliberate iterative model from an ordinary formula mistake. The examples use invented learning-project data and simple arithmetic, so a Punggol student can check each repair independently.
For a family considering how a tutor should support spreadsheet learning alongside Punggol tuition, ask to see the reasoning behind the repair. Which cells supply the result? Which dependency closes the loop? What should the correct answer be without the spreadsheet? These questions reveal understanding more clearly than a workbook that merely stops displaying a warning. The activities here are enrichment suggestions, not a claim about a specific advertised spreadsheet course.
Choose your next learning step
1. Begin with the direction of dependence
A spreadsheet formula calculates a result from referenced values. If C2 contains =A2+B2, then C2 depends on A2 and B2. Those two inputs need not depend on C2. The direction matters: the result reads the inputs.
For paper practice, draw an arrow from each input towards the result it helps calculate. Label your arrow convention explicitly, because diagrams sometimes use the opposite convention to show which cell reads another. Either convention can work if it is used consistently.
Suppose A2 contains four, B2 contains six and C2 adds them. The expected result is ten. The calculation can be completed because the inputs are already available. A student can explain where the answer comes from without first knowing the answer itself.
Now place =C2+1 in C2. The formula needs C2's value to determine C2's value. That is a direct self-dependency. It cannot be treated like the ordinary addition of two independent inputs.
This first comparison gives the learner a concrete model. A circular reference is about the structure of the calculation, not simply the presence of a complicated formula or a large workbook. Even a formula with one reference can contain a loop.
2. Recognise a direct circular reference
A direct circular reference occurs when a formula refers to the cell containing that formula. In cell B5, =B5+3 reads B5 while attempting to calculate B5. In cell D8, =SUM(D2:D8) includes D8 in the range being totalled.
The second example is especially common because the formula looks like a reasonable total. A student may select a range by dragging and accidentally include the total cell. The formula then adds the result to the inputs used to produce the result.
A quick repair for that particular total is to exclude the output cell: =SUM(D2:D7) in D8, assuming rows two through seven contain the intended six input values. The range should be chosen from the task's actual data, not simply shortened until the warning disappears.
Ask the learner which rows are observations and which row is a summary. Labels can make the distinction visible. If the summary is not clearly separated from the inputs, the workbook design invites the mistake.
Do not judge the repair solely by the absence of a warning. Check that every intended input is included exactly once and that no unintended row is included. A formula can become noncircular and still compute the wrong total.
3. Trace an indirect circular reference
An indirect circular reference involves more than one cell. Suppose B2 reads C2, C2 reads D2 and D2 reads B2. No single formula needs to contain its own address explicitly. The dependency chain still returns to its starting point.
This can be difficult for a learner who searches only for the formula's own cell name. They may inspect B2, see only C2 and conclude that there is no circular reference. The missing step is following the referenced formula onward.
Use a tiny three-cell example on paper. Write B2 as =C2+1, C2 as =D2+1 and D2 as =B2+1. Start at B2, follow each required input and notice where the chain closes.
The repair depends on the intended model. One cell might need to become a genuine input. Another formula might need to refer to a different source. Deleting an arbitrary link can remove the loop while destroying the calculation's purpose.
For a tutor, the useful question is “Which value should be known independently?” A spreadsheet often becomes circular when a learner confuses an input with a result. Recovering the intended input-output relationship is more important than finding the shortest formula edit.
4. Inspect the cell that stores the formula
To understand a formula, record both its text and its location. The formula =SUM(B2:B7) may be a sensible total in B8, but it is circular if stored in B7 because its range includes its own cell. Formula text alone does not establish the dependency structure.
This is an important lesson for students who copy a formula from a demonstration. The destination cell is part of the meaning. A formula that was correct in one position can become incorrect when pasted into a different position.
Relative references can also shift during copying. The learner should inspect the actual resulting references, not assume that the copied formula still points at the original range. Dollar signs control reference movement during copying; they do not automatically prevent circularity.
For practice, write the same total formula in two different destination cells on paper and ask whether each range includes its destination. Then identify which destination matches the workbook's design.
A parent can ask, “Where is that formula written?” alongside “What does it say?” These two questions are simple enough for a nonexpert adult and precise enough to reveal many accidental totals.
5. Repair a six-row practice total
Use six invented practice counts in B2 through B7: four, six, five, three, seven and five. Their total is thirty. Put a summary label in A8 and the intended total formula in B8.
If B8 contains =SUM(B2:B8), the range includes the summary cell and creates a direct circular reference. Correcting it to =SUM(B2:B7) makes the calculation depend only on the six inputs.
The hand calculation is a useful independent check: four plus six is ten, then five makes fifteen, three makes eighteen, seven makes twenty-five, and the final five makes thirty. The learner should be able to verify the spreadsheet's result without trusting the same formula again.
Now change one input, such as B4 from five to eight. The total should increase by three to thirty-three. This checks whether the repaired formula responds to the intended source cells. A pasted constant thirty would pass the first check but fail the second.
Finally, insert or add a new observation according to the chosen workbook layout and inspect the range again. The repair should suit the way the project will grow. A warning-free formula that silently omits future data is not a complete design solution.
6. Separate inputs, calculations and summaries
A clear workbook gives independent inputs, intermediate calculations and summaries recognisable roles. They need not live on separate worksheets, but the arrangement should help a learner see which cells are entered and which are computed.
For a small educational example, use one block for invented observations, another for calculated differences and a clearly labelled summary below or beside the data. A total should read the observation block; observations should not read the total unless the model intentionally requires that relationship.
Labels are more important than decorative colour. A pale input fill can help, but the workbook should remain understandable from headings and formulas. Colour alone does not explain whether a value is an independent assumption or a calculated result.
When a circular reference appears, use the layout to ask whether a summary has been fed back into its own source block. The formula may reveal a design problem rather than an isolated typing error.
For a Punggol student making a simple reading-project tracker, invented page counts are enough. The lesson does not need real pupil data or comparisons between children. The useful skill is organising a checkable calculation with a visible direction of dependence.
7. Use Excel's circular-reference tools
In desktop Excel, the Formulas tab includes Error Checking and a Circular References list for locating affected cell addresses. Microsoft also provides Trace Precedents and Trace Dependents tools for following formula relationships. Availability and presentation can vary by platform and version.
Start by locating the reported cell, then inspect its formula and the cells it references. The reported address is a starting point for investigation; it does not necessarily explain the entire loop by itself.
If the chain spans several cells, trace the dependency one step at a time. Keep a small written sequence of addresses so the learner can see when a previously visited cell appears again. This makes the loop explicit.
Repeat the check after a repair. A workbook can contain more than one circular reference, and fixing one may reveal another. Do not assume the first reported cell was the only problem.
For web or mobile versions where auditing commands are limited, inspect the available formulas and use an appropriate desktop version if needed. The educational goal remains the same: identify the dependency and explain the repair. A particular menu path is a tool for that investigation, not the whole lesson.
8. Distinguish precedents from dependents
A cell's precedents are the cells it uses as inputs. Its dependents are cells whose formulas use it. Knowing which direction you are tracing prevents an investigation from wandering through unrelated results.
If C2 contains =A2+B2, A2 and B2 are precedents of C2. C2 is a dependent of A2 and of B2. To understand what C2 needs before it can be calculated, follow its precedents.
To understand what might change if A2 is edited, follow its dependents. That is a different question. Both directions are useful, but a circular-reference investigation needs a consistent view of the chain.
For paper practice, choose one cell and write two sentences: “This cell reads…” and “These cells read this cell…”. The learner can then connect the vocabulary to the actual formulas.
A tutor should avoid using the terms as labels to memorise without a working example. Once the student can name the direction of dependence, the auditing tools become easier to use. The diagram and the formulas describe the same relationship.
9. Follow a loop across worksheets
A circular reference can cross worksheet boundaries. A summary sheet might read a value from a data sheet, while a formula on the data sheet reads that summary value. The loop is still present even though the formulas are not visible together.
Record the worksheet name with the cell address. A reference such as Summary!B2 identifies a different location from Data!B2. Omitting the sheet name from a tracing note can make two distinct cells look identical or hide a return to the starting cell.
Use a fictional two-sheet example. On Summary, B2 reads Data B2 plus one. On Data, B2 reads Summary B2 plus one. Follow the references and explain why neither calculated value has an independent starting point.
The correct repair may be to enter a base observation on Data and let Summary calculate from it. If the intended process is different, state that process before editing. The worksheets' names alone do not decide which value should be an input.
External workbooks and more complex references can make tracing harder. For a learner's first investigation, reduce the model to a small local example that preserves the dependency structure. Understanding the simple loop is a better foundation than clicking through a large unfamiliar workbook.
10. Do not confuse copied formulas with iteration
A formula copied down a column can represent a sequence of time steps without creating a circular reference. If B2 contains an initial value and B3 contains =B2+5, copying the formula downward lets each new row read the previous row.
The dependency chain moves from earlier rows to later rows. It does not return to the same cell. The row representing the next step is distinct from the row representing the current step.
This is useful when teaching repeated change. A learner may hear that a calculation is repeated and assume it must use circular references. Repetition in a model and circular dependence in a cell graph are not the same thing.
For an invented accumulation example, begin with ten in B2, then add five per row. The next values are fifteen, twenty and twenty-five. Each is calculated from an already available previous value.
If the learner instead writes =B3+5 in B3, that row reads itself and becomes circular. The difference is not the addition operation; it is the referenced position. Keeping time steps in separate rows can make a repeated process transparent and easy to check.
11. Know when iteration is deliberate
Some spreadsheet models deliberately use circular calculations and iterative recalculation. In those cases, the model's designer specifies how repeated calculations should proceed and what stopping conditions should apply. Enabling iteration is a modelling decision, not a general way to repair every warning.
A deliberate iterative model needs a defined equation, a starting value, an update rule and a reason to expect suitable behaviour. It should also distinguish a mathematical solution from the numerical process used to approach it.
For introductory learning, it is usually clearer to show the repeated updates in separate rows before using a circular cell. The student can inspect each step and see whether the values approach a limit, alternate or grow.
Do not enable iteration simply because Excel offers the option. If a total accidentally includes itself, the sensible repair is normally to correct the input range. Iteration does not make an accidental dependency express the intended total.
A parent evaluating a lesson can ask, “Is this loop intentional, and what equation is it solving?” If the learner cannot answer, return to the dependency investigation before changing calculation settings.
12. Solve a simple feedback equation algebraically
Consider the purely educational equation x = 100 + 0.1x. Move the feedback term to the other side: 0.9x = 100. Dividing gives x = 100 / 0.9, approximately 111.111111. The algebra identifies the fixed point.
A spreadsheet can calculate the answer directly from independent inputs. If B2 holds the coefficient 0.1 and B3 holds the base value 100, an output cell can use =B3/(1-B2). This avoids a circular dependency for this simple linear model.
The denominator must not be zero. If the coefficient equals one, that direct formula cannot provide a finite answer by division. The original equation must be analysed rather than treated as an ordinary input case.
The example is a toy mathematical model, not a financial estimate or a recommendation for a real budgeting calculation. Its purpose is to compare a fixed-point equation with a noncircular algebraic solution.
Ask the learner to substitute the approximate answer back into the equation. The check will reflect rounding. A direct formula can be independently assessed through the original relationship, which is stronger than merely accepting the displayed value.
13. Show convergence in separate rows
For the same equation, define the update x_next = 100 + 0.1x_current and begin at zero. The first updates are 100, 110, 111 and 111.1. The sequence approaches the fixed point because the error is multiplied by 0.1 at each step.
In a spreadsheet, put zero in B2 and =100+0.1*B2 in B3, then copy the update down. Each row represents a separate step. The formulas are not circular because each reads the previous row.
Add a column for the change between consecutive values. This makes the approach visible and gives the learner a quantity to inspect. A decreasing change supports the observed behaviour in this example, while the algebra explains why it occurs.
For an update of the form x_next = b + r*x_current, the simple scalar iteration converges to its fixed point from ordinary starting values when the absolute value of r is below one. That is a property of this specific linear iteration, not a general guarantee for every spreadsheet model.
The educational benefit is seeing the model and the numerical process separately. The fixed point is one question; whether the chosen update reaches it is another.
14. See why a fixed point can exist without convergence
Change the update to x_next = 100 + 1.1x_current. The algebraic equation still has a fixed point: solving x = 100 + 1.1x gives negative one thousand. But starting the iteration at zero does not approach that fixed point.
The first updates are 100, 210 and 331. The values grow away from the fixed point because the error is multiplied by 1.1 rather than a factor with absolute value below one.
This example prevents a common misunderstanding: an equation having a solution does not mean an arbitrary repeated-update method will converge to it. Numerical behaviour needs its own check.
A learner who enables spreadsheet iteration and sees a number should therefore ask how it was obtained and whether the process behaves as intended. A displayed result may be a stopped intermediate value rather than an established solution.
For the lesson, keep the example in separate rows so the pattern is visible. The student can compare the convergent and divergent updates side by side without relying on hidden recalculation settings.
15. Recognise oscillation rather than progress
Consider x_next = 100 - x_current beginning at zero. The sequence alternates between zero and one hundred. The equation's fixed point is fifty, but the chosen starting value does not lead the update towards fifty.
This is another reason a finite number of iterations does not establish correctness. If the process stops after one count, it may display zero; after another, it may display one hundred. Neither output is the fixed point.
A spreadsheet column showing the updates makes the alternation obvious. A single circular output cell can make the same behaviour harder for a beginner to inspect. Visible steps are especially useful during learning.
Ask the learner to calculate the fixed point algebraically and compare it with the generated sequence. Then change the starting value to fifty and observe that it stays fifty. This demonstrates that starting at a fixed point and converging to it are different situations.
The lesson remains a simple mathematical example. Real iterative models may have more complicated behaviour and require analysis appropriate to their equations. Do not generalise one toy pattern into a universal modelling rule.
16. Understand stopping settings as numerical controls
Iterative calculation settings can limit the number of recalculations and define a change threshold. These settings control when the numerical process stops. They do not by themselves prove that the model is correct or that the process converged to the intended answer.
A maximum-iteration limit can stop a divergent or slowly convergent process before its purpose is fulfilled. A small change threshold can be useful for a well-understood convergent process, but its meaning depends on the scale and units of the model.
The learner should document the stopping rule in ordinary language. “Stop when the change between updates is below this amount, or when this maximum count is reached” is clearer than “turn on the setting that makes the error disappear.”
Also record whether the process stopped because the change condition was met or because the count limit was reached, when that information can be obtained from the chosen model. Those outcomes have different implications.
For an introductory Punggol enrichment lesson, a visible row-by-row model often makes these ideas easier to teach than an actual circular workbook. The student can count the steps and inspect the changes directly.
17. Keep displayed rounding separate from calculation
A cell can display a rounded value while retaining more precision for calculation. Seeing the same displayed number in several successive rows does not necessarily mean the underlying values are identical or that an iteration has reached an exact fixed point.
For the convergent example approaching 111.111111, formatting to one decimal place may display 111.1 across several later rows. The remaining changes are simply smaller than the display resolution. Inspect more decimal places or a separate change column when the distinction matters.
Do not change precision settings casually to make the display match an expected answer. Calculation precision and formatting are different controls, and some precision settings can change stored numerical information.
For ordinary educational examples, preserve the calculated values and choose a readable display. State whether an answer is exact or approximate. A rounded display is a communication choice, not proof of numerical completion.
A parent can ask, “Are the values equal, or do they just look equal at this format?” This is a useful bridge between mathematical approximation and spreadsheet behaviour.
18. Check a repair with an independent calculation
After removing an accidental circular reference, compute at least one result independently. Use a hand total, a separate simple equation or a deliberately small data set. The check should not repeat the same faulty reference structure.
If a total of six values should be thirty, write the addition on paper. If a feedback equation has a direct algebraic solution, substitute the output into the original equation. If a copied formula should increase by a fixed amount per row, inspect the first few steps manually.
Then change one input and predict the direction and size of the result's change. This is particularly helpful for catching a repair that replaced a formula with a constant. The initial value may look correct while the model has stopped responding.
Check an edge case suited to the task. A total with all zero inputs should be zero. A direct feedback formula with a zero coefficient should return the base value. A formula intended for a fixed number of observations should make that requirement visible.
Verification is part of the lesson, not a separate punishment after making a mistake. It gives the learner a way to trust the repaired workbook for a stated reason.
19. Avoid repairs that erase the model
Replacing a circular formula with a typed value can remove the warning, but it may also remove the calculation the workbook was meant to perform. Deleting a referenced row can do the same. A clean-looking workbook is not necessarily a correct workbook.
Before editing, state the intended relationship. “This cell totals the six observation rows” or “This output solves the stated linear equation” gives the repair a purpose. The formula should express that purpose after the change.
If the intended relationship is unclear, investigate the labels and the source of the inputs. Ask the teacher, tutor or project owner for the relevant specification rather than guessing at a meaningful number.
Keep a copy of the original small example when experimenting. That lets the learner compare the faulty and repaired structures and explain exactly which dependency changed. This is a practical learning safeguard, not a reason to preserve every accidental formula indefinitely.
For a family, the useful progress statement is “The learner corrected the input range and verified the total,” rather than “The warning vanished.” The first describes understanding; the second describes only a screen state.
20. Use named ranges and table references with understanding
Named ranges and structured table references can make formulas more readable, but they do not automatically prevent circular references. If a named range includes the output cell, or a table formula depends back on its result, the loop still exists.
A readable name such as PracticeCounts should refer to the intended input region. Inspect its definition before relying on it. A clear name attached to the wrong range is still a wrong reference.
Structured references can help a growing data table include intended rows consistently. The learner should understand whether a reference addresses data rows, a particular row or a summary region. The formula's actual scope matters.
For introductory practice, use simple cell references first, then compare them with a named input range. Ask the student to explain which cells each version reads. This keeps the abstraction connected to the dependency structure.
Do not add names or tables solely to make the workbook appear advanced. Use them when they make the project easier to understand and maintain. The underlying question remains unchanged: does the calculation depend on an independent input, or does it eventually return to itself?
21. A practical investigation workshop
Case one: B8 totals the values in B2 through B7 but contains =SUM(B2:B8). Identify the direct loop, correct the range and check the total by hand. Then change one observation and predict the new total.
Case two: C2 contains =D2+2 and D2 contains =C2-2. Trace the two-cell loop. Notice that the relationships do not determine one unique pair of values by themselves. Choose an independent input only if that choice matches the intended task.
Case three: B2 holds ten and B3 contains =B2+5, copied downward. Explain why this is not circular. Each step depends on an earlier available row, and the chain does not return to itself.
Case four: an output should satisfy x=100+0.1x. Solve the equation directly, then build a row-by-row iteration from zero and compare the approach to the algebraic solution. Explain why the visible iteration can be noncircular.
Case five: change the iterative coefficient to 1.1. Predict whether the updates from zero approach the fixed point. Verify the first three values and explain the difference between a solution existing and a particular numerical method reaching it.
The workshop checks several distinct ideas. A learner can be strong at accidental totals while still needing support with indirect dependencies or numerical iteration. Use the result to choose the next lesson, rather than label the student globally as good or poor at spreadsheets.
22. Build a small Punggol learning project
Create an imaginary reading or practice project with six observations, a total and a short written interpretation. The figures can be invented openly for learning. The project does not need actual family schedules, pupil results or personal records.
Begin by entering the observations and calculating a hand total. Add the spreadsheet formula only after the expected relationship is clear. Then introduce an intentional self-including total and ask the learner to diagnose it.
In a second session, add a simple calculated column that depends on the observations. Trace the inputs and outputs. Keep the workbook small enough that the learner can describe every formula without searching through many worksheets.
For an extension, build the row-by-row fixed-point example on a separate sheet with clear labels for current value, next value and change. Do not mix the toy iteration into the ordinary practice total.
Around schoolwork, CCA and travel, the family can stop after one complete investigation. The milestone is a learner who identifies a loop, explains its cause, repairs the intended relationship and verifies the result. That is substantial progress even in a small workbook.
23. Questions parents and students often ask
Does a circular warning mean the whole workbook is unusable? It identifies a calculation issue that needs investigation. The affected dependency chain and the workbook's purpose determine the impact. Do not assume every other cell is wrong, and do not ignore the warning because some values look sensible.
Can I fix it by moving the formula? Sometimes, particularly when a total accidentally sits inside its own range. But the moved formula must still read the intended inputs, and any dependent cells must still refer to the correct output. Check the model after moving it.
Do dollar signs prevent circular references? No. They control how references change when a formula is copied. A fixed reference can still point to the formula's own cell or participate in an indirect loop.
Is iterative calculation always wrong? No. It can be appropriate for a deliberately designed model with understood numerical behaviour. It is not a substitute for correcting an accidental dependency.
What should I show a tutor? Provide the formula's location, its text, the relevant referenced cells, the intended relationship and the result expected from a small independent check. This makes the support question precise and actionable.
24. A conversation that makes the mistake useful
When a warning appears, begin with “Which value is this formula trying to calculate?” Then ask where its inputs come from. This keeps the conversation about the calculation rather than about the student's competence.
If the student can identify a direct self-reference, ask whether the intended range includes the output. If the loop is indirect, follow one dependency at a time and keep the addresses visible. If the model is intentionally iterative, ask for the equation and the stopping rule.
Let the learner describe the repair before making it. A sentence such as “The total should read only the six observation cells” is a useful plan. After the edit, the student can check whether the formula now expresses that sentence.
A short correction note can preserve the learning: the original formula included its own cell, the intended input range ended one row earlier, and the repaired total passed a hand calculation and an input-change test. That is enough to make the explanation reusable.
For Punggol parents, this is a practical way to support digital learning without taking over the workbook. The adult can ask for a clear relationship and a check, while the learner remains responsible for understanding the calculation.
25. Continue through the right reference
Microsoft's circular-reference guidance explains how to locate a circular dependency and the distinction between removing a loop and allowing deliberate iteration. Its recalculation and iteration guidance addresses the calculation controls.
The existing eduKate guide to creating spreadsheets with Super Intelligence provides a broader context for formulas, models and checking. This lesson's narrower job is tracing and repairing a circular dependency rather than generating a complete workbook.
For the family learning arrangement, return to How Studying Works at eduKatePunggol. Bring the learner's actual attempted formula and a small reproducible example to the next support conversation.
Spreadsheet circular references stop being mysterious when the student can follow the dependency and explain where it closes. A good repair restores the intended relationship and passes an independent check. That gives the learner something more valuable than a warning-free screen: a calculation they understand well enough to change responsibly.
26. Check additions and deletions after a repair
A repaired total should remain understandable when the input block changes. If a student adds an observation, inspect whether the formula includes the new row. If they delete an observation, inspect whether the range still addresses the intended remaining inputs. The correct behaviour depends on how the workbook is designed to grow.
Use a small example with a clearly separate summary row. Add a seventh invented observation of four to the original total of thirty. The expected result is thirty-four. Then remove that added observation and confirm that the total returns to thirty. These changes test whether the formula follows the data rather than a remembered answer.
If a table or named range is used, explain how it expands or remains fixed in this project. Do not assume that a familiar-looking name automatically includes every future row. Inspect the actual referenced region and verify the resulting formula.
The point is to connect a local repair with the workbook's next ordinary use. A calculation can be correct for today's six inputs while silently omitting tomorrow's seventh. That is a different failure from circularity, but a careful repair should make the intended data boundary clear.
27. Write a dependency explanation another person can check
Ask the learner to write a short account containing the output cell, the intended input cells, the accidental return path and the repaired formula. Keep the explanation tied to the actual workbook, including worksheet names where needed.
For the six-row total, the note can say that B8 is a summary, B2 through B7 are independent observations, and including B8 in the sum makes the summary depend on itself. The corrected formula reads only the observations. The hand total and the input-change test provide independent evidence for the repair.
For an indirect loop, list each step until the first repeated location appears. If Summary B2 reads Data B2 and Data B2 reads Summary B2, the repetition identifies the closed chain. The learner should then state which value should be supplied independently under the intended model.
For an iterative extension, state the equation, the starting value, the update rule and the stopping condition separately. This keeps a deliberate mathematical process distinguishable from an accidental self-including total.
The explanation need not be long. Its purpose is to let a tutor, teacher or classmate reconstruct the reasoning without guessing at hidden assumptions. A student who can write that account is showing control of the dependency structure, not just familiarity with a menu command.
At the next session, provide a different tiny workbook and ask for the same style of explanation. The learner can then demonstrate that the method transfers: locate the result, follow its inputs, identify any return path, repair the intended relationship and check the outcome.
