How do you investigate the cause of the main pain point?

You know where the target group gets stuck: proposals are difficult to compare and a recommendation remains unfinished. But why does this happen? Are shared selection criteria missing, do suppliers provide different information, or is it unclear who has authority to decide?

The Cause of the main pain point field requires you to go beyond the complaint. Describe how the problem arises or persists. This makes it easier to formulate focused research questions and possible improvements. Quality lies in the evidence for the explanation, not in how convincing the story sounds.

In the target group profile, this field describes the factor or combination of factors that causes or sustains the main pain point, explaining how it works and what evidence is available. When the cause has not been established, the field contains an explicit hypothesis.

This is Neurofactor’s practical working definition. The field name is singular, but that does not mean every problem has exactly one cause. Clarify what was observed, which explanation is proposed and what remains to be investigated.

Use the distinctions below to keep different kinds of information in their proper place. The examples are fictional.

A factor’s role depends on the research question. Time pressure may be a trigger, but could also contribute to less careful comparison. Explain its role rather than simply assigning a label. A cause also need not be sufficient on its own to produce the problem.

Cause is a general scientific concept, not a marketing term with a single identifiable originator. Three publications help inform this profile field.

Assessing associations. In 1965, Hill discussed viewpoints for assessing causation, including temporal order. He did not present them as conclusive rules of proof. Hill, 1965.

Making assumptions explicit. Pearl shows that statistical association alone does not substantiate a causal conclusion: causal assumptions are also required. Pearl, 2009.

Handling self-explanations carefully. Nisbett and Wilson discuss limitations of introspective explanations of mental processes. A stated reason therefore is not automatically an established cause; this does not mean every self-report is wrong. Nisbett & Wilson, 1977.

These publications come from different research fields. The application to the target group profile below is Neurofactor’s own; the sources do not validate a fixed template or automatic cause scores.

“Lack of trust” sounds like a cause, but leaves much unspecified. Trust in what, arising from which experience, and with what consequence for the task? An explanation becomes more useful when you identify the intermediate steps.

A fictional hypothesis could be: suppliers use different definitions of reach; the evaluator therefore cannot make a like-for-like comparison; the recommendation remains unfinished. This describes a possible mechanism, not proof that the chain actually occurred.

Ask which observation would fit the explanation and which would make it less plausible. If recommendations also stall when definitions match, investigate further. Decision authority or budget may be missing, or several factors may be involved.

A clear explanation need not identify the deepest imaginable “root cause”. Choose a scope suited to the task and research question. State when you are investigating a contributing factor without claiming that it explains everything.

The approach below is a practical framework for research and writing.

1. Fix the problem statement. Use the concrete sentence from the preceding profile field. Do not change the complaint along the way to fit a favourite explanation.
2. Reconstruct the event. Ask what happened before, during and after the difficulty. Where possible, use documents, versions, timestamps or observations alongside the interview.
3. Formulate several explanations. Record missing information, different assessment criteria or unclear authority as separate hypotheses, for example.
4. Describe the mechanism for each hypothesis. Which intermediate process would explain the problem? What evidence would you expect to find?
5. Seek data that distinguish the explanations. Compare cases where the problem occurs with cases where it does not. Investigate other differences that might explain an outcome. One favourable case does not settle the question.
6. Investigate a targeted change where possible. Specify in advance what will change, what the comparison will be and which outcome you will measure. Limit simultaneous changes and use an appropriate controlled design where feasible.
7. Match the conclusion to the evidence. State the factor, mechanism, context and uncertainty. Retain evidence against the preferred explanation as well.

Ask, for example: “What information was missing at that moment?”, “What could you not do as a result?”, “When did comparison work?” and “What changed then?” Repeatedly asking “why?” is not an independent test of the answers.

Use appropriate wording. The labels below are editorial aids, not a validated evidence hierarchy.

A controlled study can support an effect of a change without proving every intermediate step. A well-supported explanation for one organisation also does not automatically apply to an entire segment. Choose wording based on what the actual design permits.

Fictional illustration, not a research finding. The main problem from the previous article is that insufficient comparability between proposals obstructs a substantiated recommendation. We now investigate three possible explanations.

Suppose the proposal files in this fictional case do indeed contain different units. That supports the existence of a comparison problem. It does not yet establish that this is the sole or decisive cause of the unfinished recommendation.

A provisional profile sentence could read:

> Different definitions and units in proposals may contribute to the comparison stalling because the evaluator needs additional information first.

You could then investigate whether equivalent information improves the comparison while controlling other differences as well as possible. If decision authority changes at the same time, do not attribute improvement exclusively to the proposal information.

Associations can be part of an explanation to investigate. An associative target group is a defined group of people whose similar associations guide decisions within a specific choice context. That does not yet establish that a particular association causes the main problem.

For example, investigate the hypothesis that “unfamiliar supplier” is associated with “difficult to justify”, leading to requests for more evidence. Keep three questions distinct: does the association occur, is it related to the trade-off, and has its causal contribution been investigated? This example is fictional.

Use an association measure or EEG outcome only for conclusions supported by the research design. A measured response does not automatically explain a choice. Keep practical factors, such as budget and procedure, in view as well; the cause need not be exclusively psychological.

• Repeating the problem. “Comparison fails because proposals are incomparable” does not explain how the barrier arises.
• Entering a character judgement. “The target group is indecisive” calls for concrete situations and research into circumstances.
• Treating a consequence as a cause. Extra meetings may arise because comparison stalls; check the temporal order.
• Collecting confirmation only. Look for situations that do not fit your explanation too.
• Reasoning backwards from a solution to necessity. Offering a comparison service does not show that it removes the cause.
• Jumping from improvement to causation. Record what else happened while the change was introduced.

For an explanation still to be investigated:

> [Factor] may contribute to [pain point] in [context] through [proposed mechanism].

For a supported conclusion:

> Within [studied context], [data] support a contribution of [factor] to [pain point] through [supported mechanism].

Use the second template only if the stated mechanism is also supported; otherwise retain it as a hypothesis. Keep source references, research design, alternative explanations, scope and open questions alongside the sentence. If no data exist yet, enter “Cause not yet researched” and note which hypotheses you intend to test.

This field connects the main problem with a factor and a mechanism you can investigate. A concrete hypothesis with a clear evidence status is more useful than an assertive explanation without support. Make visible which alternatives have been investigated and which questions remain open.

Want to investigate why your target group gets stuck and which associations play a role? Discuss the choice context, available evidence and open research questions with Neurofactor.