When priorities compete

01Choices rely on intuition rather than a consistent method.

02Too many criteria make priorities difficult to compare.

03Inconsistent judgments can undermine the final ranking.

04Different stakeholder priorities make consensus difficult.

AHP derives weights from pairwise comparisons — and checks consistency.

FROM SETUP TO FIELDWORK

Build the hierarchy. See the exact comparison respondents make.

No separate preview step — every criterion you add becomes a pairwise comparison on the right immediately.

createsurvey · ahp-analysis

AHP Priority Analysis

PriorityCustom tasks · min. 20 respondents

Decision goal

Select the best option

What every comparison ultimately decides. Respondents are shown each pair with the standard: "which matters more toward [this goal]?"

Criteria hierarchy (up to 5 levels)

Each criterion is one factor that matters to the decision (e.g. "Price"). Add at least 2. Use "Sub" to split a criterion further — alternatives are only ever compared under leaf (bottom-level) criteria.

1Price+ Sub×

2Quality+ Sub×

3Delivery speed+ Sub×

+ Add criterion

Alternatives

Apply to survey

Live

Live preview

Question 1 of 4

Goal: Select the best option

Price

Quality

← PriceEqualQuality →

‹ Previous

* Required

Next ›

Left: the criteria hierarchy every AHP study starts from. Right: exactly what respondents see — one pairwise comparison at a time, updated as you type.

Hierarchy structure

AHP structures a single goal into criteria and alternatives, then compares them layer by layer. Sub-criteria can split further, so the hierarchy from goal to alternatives can run up to 5 levels deep.

GOAL

One decision, stated once.

Every hierarchy starts from a single stated goal — everything below it exists to answer that one question.

CRITERIA

Calculate criterion weights.

Pairwise comparisons convert judgments into ranked weights for each first-tier criterion.

SUB-CRITERIA

Criteria split as far as you need.

Any top criterion can branch into finer sub-criteria, and each of those can branch again — up to 5 levels from goal to alternatives.

ALTERNATIVES

The tree ends in one winner.

Every alternative is scored against the full weighted hierarchy so the final ranking reflects all criteria.

FROM RESPONSES TO RESULTS

From pairwise comparison to final ranking

Every figure below comes from one real example study — a laptop choice with 30 respondents, 5 criteria and 3 alternatives, computed by the same engine your own survey runs on.

Evaluation structure

Check the rating scale, the comparison method, the criteria structure and the response status.

Study designSaaty 1–9

MethodPairwise (Saaty 1–9)

AggregationGeometric mean

Criteria5 top · 7 leaf

Alternatives3

Valid responses30 / 30

1equal3moderate5strong7very strong9extreme

Criterion weights

See the importance of each criterion as a weight.

Criterion weightsTotal 100%

Brand / support

39.1%

Price

20.7%

Battery life

17.6%

Build quality

13.3%

Performance

9.3%

Interpretation

Brand / support is the top criterion at 39.1% of the decision.

Sub-criterion weights

See importance within each parent criterion together with overall influence.

Sub-criteria7 leaves

Purchase cost← Price80.0% · 16.6%

Design← Build quality75.0% · 10.0%

Graphics← Performance42.9% · 4.0%

Running cost← Price20.0% · 4.1%

* Local weight within its parent, then the global weight multiplied down the hierarchy.

Judgment consistency

The consistency ratio (CR) verifies how reliable the comparison results are.

ConsistencyCR < 0.1

MatrixnλmaxCR

Top criteria54.758−0.0540

Performance33.0000.0000

Battery — alts33.0740.0634

Matrices under CR < 0.113 / 13

Per-respondent check: 0 / 30 above CR 0.1.

Final ranking

Criterion weights and alternative evaluations are combined to compute the ranking.

Final priorities3 alternatives

1· ThinkPad X10.411

2· Dell XPS0.365

3· MacBook Air0.224

* Score = Σ(criterion weight × local priority).

Winner

ThinkPad X1

Brand / support0.600 (1st)

Graphics0.648 (1st)

Purchase cost0.356 (2nd)

Battery life0.117 (3rd)

Ranking stability

See how the ranking changes when the importance of a criterion shifts.

Sensitivity5 criteria

Price

21%

Performance

Battery life

18%

Build quality

13%

Brand / support

39%

Now

ThinkPad X1 0.41

Dell XPS 0.37

MacBook Air 0.22

Drag a weight and the ranking below re-sorts immediately.

A COMPLETE REPORT

See the full report from summary to conclusion.

The six cards above are fragments of this. Scroll the real output end to end, right here on the page.

AHP example — laptop choice · 30 simulated respondents

Analytic Hierarchy Process (AHP)

Pairwise comparison analysis — criterion weights, consistency ratios, sensitivity, and alternative scoring for "Best laptop choice".

Goal

Best laptop choice

Criteria

9 leaf nodes

Respondents

with AHP answers

Consistency Ratio

-0.0540

consistent

Analysis Settings

Adjust the weight-derivation method and consistency filters used in the report below.

Weight derivation

CR filter threshold

Outlier detection (MAD)

Executive Summary

30 respondents · 5 criteria

This AHP study combined pairwise judgments from 30 respondents across 5 top-level criteria and 7 sub-criteria and 3 alternatives, using the geometric mean method. Brand/support is the top-weighted criterion at 39.1%, and overall judgment consistency is CR = -0.0540 (acceptable, < 0.1). The leading alternative is "ThinkPad X1" with a global priority score of 41.1%, 4.6 points ahead of "Dell XPS".

Study overview

Metric	Value
Respondents	30
Top-level criteria	5 — Brand/support · Price · Battery life · Build quality · Performance
Sub-criteria	7
Alternatives compared	3 — MacBook Air · Dell XPS · ThinkPad X1
Comparisons per respondent	42
Comparison answers in total	1,260
How it was answered	Two items compared and rated 1–9
How answers were combined	Answers pooled into a weight per criterion
How the range was computed	Respondents resampled 1,000 times

The evaluation structure

The goal was split into top-level and sub-criteria, and the alternatives were compared at the end. At each level, only items of the same level were compared with each other.

What was judged most importantsums to 100% per tier

Each criterion's weight. The range in brackets is how far the result can move when the respondents change.

Top-level criteria

Brand/support39.1%(33.8%–41%)

Price20.7%(19.6%–24.1%)

Battery life17.6%(16.5%–20.8%)

Build quality13.3%(12.8%–16.9%)

Performance9.3%(8.1%–10.9%)

Interpretation

Brand/support came out as the most important criterion in 100.0% of the draws when the respondents were resampled. The ranges for Price·Battery life, Battery life·Build quality overlap, so those criteria cannot be put in a clear order.

Sub-criteria

Purchase costunder Price

80.0% of parent · 16.6% total

Designunder Build quality

75.0% of parent · 10.0% total

Maintenance costunder Price

20.0% of parent · 4.1% total

Graphics performanceunder Performance

42.9% of parent · 4.0% total

Memory/storageunder Performance

42.9% of parent · 4.0% total

Durabilityunder Build quality

25.0% of parent · 3.3% total

CPU performanceunder Performance

14.3% of parent · 1.3% total

※ "of parent" is the local weight within that criterion alone; "total" is the composite weight once propagated down the whole hierarchy.

Comparison answers by criterion

How much more important the row criterion was said to be than the column criterion. Above 1 means the row was judged more important.

Top-level criteria

CR -0.0540

	Price	Performance	Battery life	Build quality	Brand/support
Price	1	3.00	1.00	1.00	1/2.0
Performance	1/3.0	1	1/2.0	1/2.0	1/3.0
Battery life	1.00	2.00	1	1.00	1/3.0
Build quality	1/2.0	1.00	1.00	1	1/3.0
Brand/support	2.00	3.00	2.00	3.00	1

Price

CR 0.0000

	Purchase cost	Maintenance cost
Purchase cost	1	4.00
Maintenance cost	1/4.0	1

Performance

CR 0.0000

	CPU performance	Graphics performance	Memory/storage
CPU performance	1	1/3.0	1/3.0
Graphics performance	3.00	1	1.00
Memory/storage	3.00	1.00	1

Build quality

CR 0.0000

	Durability	Design
Durability	1	1/3.0
Design	3.00	1

1 Equal3 Moderate5 Strong7 Very strong9 Extreme(2·4·6·8 are intermediate values, Saaty 1–9 scale)

Matrix	n	λmax	CI	CR	Status
Top-level criteria	5	4.758	-0.0605	-0.0540	OK
Price	2	2.000	0.0000	0.0000	OK
Performance	3	3.000	0.0000	0.0000	OK
Build quality	2	2.000	0.0000	0.0000	OK
Brand/support — alternatives	3	3.000	0.0000	0.0000	OK
Battery life — alternatives	3	3.074	0.0368	0.0634	OK
Purchase cost — alternatives	3	2.716	-0.1418	-0.2445	OK
Design — alternatives	3	3.000	0.0000	0.0000	OK
Maintenance cost — alternatives	3	3.018	0.0091	0.0158	OK
Graphics performance — alternatives	3	3.004	0.0018	0.0032	OK
Memory/storage — alternatives	3	3.000	0.0000	0.0000	OK
Durability — alternatives	3	3.000	0.0000	0.0000	OK
CPU performance — alternatives	3	3.000	0.0000	0.0000	OK

✓ All pairwise matrices meet the CR < 0.1 benchmark — judgments are internally consistent.

Weight of each sub-criterion

Each sub-criterion's weight, split into its share within its own top-level criterion and its share of the whole.

Top-level criterion	Sub-criterion	Within its criterion	Of the whole
Price 20.7%	Purchase cost	80.0%	16.6%
Build quality 13.3%	Design	75.0%	10.0%
Price 20.7%	Maintenance cost	20.0%	4.1%
Performance 9.3%	Graphics performance	42.9%	4.0%
Performance 9.3%	Memory/storage	42.9%	4.0%
Build quality 13.3%	Durability	25.0%	3.3%
Performance 9.3%	CPU performance	14.3%	1.3%

Dell XPS

0.333

ThinkPad X1

0.333

FINAL RECOMMENDATION

ThinkPad X1

Brand/support0.600 (#1)

Battery life0.117 (#3)

Purchase cost0.356 (#2)

Design0.200 (#2)

Maintenance cost0.210 (#3)

Graphics performance0.648 (#1)

Memory/storage0.600 (#1)

Durability0.429 (#1)

CPU performance0.333 (#1)

0.411Global priority score (highest among alternatives)

Global Priority Ranking

#1 · ThinkPad X1first in 99.3%0.411

#2 · Dell XPSfirst in 0.7%0.365

#3 · MacBook Airfirst in 0.0%0.224

※ Global score = Σ(criterion weight × local priority under that criterion), summed across all criteria.

Interpretation

Under the criteria as they stand, ThinkPad X1 is first at 41.1% and Dell XPS second at 36.5%. The gap is 4.6 points, and ThinkPad X1 was still first in 99.3% of the draws when the respondents were resampled. Changing the criterion weights can change the ranking, so the shifts can also be checked below.

Consistency of the comparison answers

Whether one respondent's comparisons fit together. For example, rating A twice as important as B and B twice as important as C implies answering that A is about four times as important as C.

All answers pooled

-0.054

Computed from the pooled answers.

Median per respondent

0.044

The middle of the per-respondent values.

Respondents whose answers disagree

0.0% of 30 · highest 0.077

Per respondent

0.1 — the line commonly used

Interpretation

Pooling every answer gives -0.054, while 0 respondent(s) answered comparisons that disagree with each other. Read the two together. The respondents and comparisons to check are listed below.

Priorities by respondent

Which criterion each respondent put first, and how similar their priorities were overall.

Agreement between respondents

0.885

Closer to 0 means the orderings differ more; closer to 1 means they are alike.

Criteria chosen first

1 criteria were put first across 28 responses.

The criterion named most important

Brand/support

28 · 100.0%

Price

0 · 0.0%

Performance

0 · 0.0%

Battery life

0 · 0.0%

Build quality

0 · 0.0%

Try changing the criterion weights

Adjust how important each criterion is and see how the ranking of alternatives changes. Moving a slider updates the scores and the ranking for the adjusted weights straight away.

Where the leader changes

Price 20.7%above 40.5% → Dell XPS leads

Performance 9.3%the leader never changes

Battery life 17.6%above 25.0% → Dell XPS leads

Build quality 13.3%above 28.0% → Dell XPS leads

Brand/support 39.1%below 31.0% → Dell XPS leads

Price21%

Performance9%

Battery life18%

Build quality13%

Brand/support39%

Current

ThinkPad X1 #1 0.41

Dell XPS #2 0.37

MacBook Air #3 0.22

Drag a bar above to test how the ranking responds to a different weight on that criterion.

Conclusions and next steps

Which alternative leads under the criteria as they stand, and what to check before deciding.

Under the criteria as they stand, the alternative to consider first is ThinkPad X1 (41.1%). It also scored highest in 99.3% of the draws when the respondents were resampled.

The heaviest item was Brand/support (39.1%), and among the sub-criteria Purchase cost (16.6%) carried the largest share.

ThinkPad X1 and Dell XPS are 4.6 points apart. The ranking can change with how heavily each criterion is weighted, so check the weight on Battery life first.

Respondents named these criteria most important: Brand/support 28. Discuss that difference before sharing the overall result or settling an action plan.

Methodology — Criterion weights were derived from pairwise comparison matrices via Saaty's AHP (1980), using the eigenvector or geometric-mean method. Consistency Ratio (CR) = CI / RI, where CI = (λmax − n) / (n − 1). Global priority scores use hierarchical composition (weighted sum) across the criteria tree.

Scroll to explore the full report

IN THE REPORT

Defensible priorities

Compare options in pairs, validate consistency, and turn stakeholder input into a ranking you can explain.

Pairwise matrix & weights

Every 1:1 comparison you make gets folded into a single consistency-checked weight per criterion — no manual spreadsheet math.

Up to 5 hierarchy levels

Model goal, criteria, sub-criteria, and alternatives as one tree — sub-criteria can branch again, up to 5 levels deep.

Consistency ratio (CR)

A consistency ratio is calculated for every comparison matrix, and matrices above the CR threshold are flagged for review.

Agreement & concentration

See at a glance how tightly 30 respondents agreed, and whether the weight is concentrated in one criterion or spread evenly.

Segment AHP

Compare priorities across age, department, role, or other respondent groups instead of relying only on the pooled average.

Final ranking & radar

The weighted score across every criterion becomes one ranked list and a radar profile, so you see the whole shape of the decision at once.

Group AHP

Aggregate many evaluators' judgments — by geometric mean or weighted average — into a single defensible priority the whole group can stand behind.

Auto interpretation

Every report closes in plain language — no need to read raw weight tables to know what the numbers actually mean.

ALSO IN SURVEY ANALYSIS

More survey methods

Each survey analysis method answers a different question — same interface, different lens.

OPP · IDP · PSM

Van Westendorp PSM

Turn four price questions into OPP, IDP, elasticity, revenue simulation, and competitor benchmark.

WTP · Part-worth

CBC Conjoint

Measure attribute importance and willingness to pay with choice-based conjoint analysis.

Kano · NPS

Kano Model

Classify features into Must-be, Performance, Attractive, and Indifferent categories.

IPA · Regression

Driver Analysis

Find which attributes drive overall satisfaction or NPS using importance-performance analysis.

Recall · Image

Brand Awareness

Track unaided and aided brand recall, brand image dimensions, and competitive positioning.

eNPS · Pulse

Employee Engagement

Measure and benchmark employee engagement, satisfaction, and eNPS across teams.

Set priorities with evidence

Turn stakeholder judgments into weighted priorities, consistency checks, and a final ranking your team can stand behind.

Get started See pricing

No credit card requiredUp to 5 hierarchy levels

Priorities you can defend.

When priorities compete

Build the hierarchy. See the exact comparison respondents make.

From pairwise comparison to final ranking

Evaluation structure

Criterion weights

Sub-criterion weights

Judgment consistency

Final ranking

Ranking stability

See the full report from summary to conclusion.

Analysis Settings

The evaluation structure

What was judged most importantsums to 100% per tier

Comparison answers by criterion

Top-level criteria

Price

Performance

Build quality

Weight of each sub-criterion

Overall score by alternative

Brand/support

Battery life

Purchase cost

Design

Maintenance cost

Graphics performance

Memory/storage

Durability

CPU performance

ThinkPad X1

Global Priority Ranking

Consistency of the comparison answers

Priorities by respondent

Try changing the criterion weights

Conclusions and next steps

Defensible priorities

More survey methods

Van Westendorp PSM

CBC Conjoint

Kano Model

Driver Analysis

Brand Awareness

Employee Engagement

Set priorities with evidence

Analysis Settings

The evaluation structure

What was judged most importantsums to 100% per tier

Comparison answers by criterion

Top-level criteria

Price

Performance

Build quality

Weight of each sub-criterion

Overall score by alternative

Brand/support

Battery life

Purchase cost

Design

Maintenance cost

Graphics performance

Memory/storage

Durability

CPU performance

ThinkPad X1

Global Priority Ranking

Consistency of the comparison answers

Priorities by respondent

Try changing the criterion weights

Conclusions and next steps

Priorities
you can defend.