Market, Competition, and Customer Research

You accelerate research without mixing source quality, method, and interpretation.

After this chapterYou can conduct reproducible research and translate it into a decision memo with evidence and uncertainty.

Your progress0 of 48 lessons

4.1

A business research brief

A research assignment begins with the decision that the evidence must support.

Formulate decision question, scope, target group, geography, period, definitions and evidence threshold. Split unknowns into sub-questions and determine suitable primary, official and secondary sources.

Note in advance what the research cannot prove and who checks interpretations. Deep research and web research help, but do not replace source evaluation.

Decision
Scope
Definitions
Source plan
Evidence threshold
Limitation

Terms in plain language

Scope: What the task includes and excludes.

How you can use this

A market study first defines what 'SME' and 'potential customer' mean in Belgium.

Try this prompt

Write a research brief for [decision] with scope, definitions, sub-questions, source hierarchy, exclusions, evidence threshold, and decision note.

Knowledge check

A business owner asks you to “research the AI market”. The actual decision concerns a trial service for Belgian maintenance companies with small administrative teams. What do you establish first?

An overview of promising AI applications, pricing models and successful Belgian services.A decision question with the segment, problem, region, period and evidence needed for this trial service.A list of AI suppliers for maintenance companies, compared on features and costs.

Your practical exercise

Write the research brief for Noor’s question of whether expanding the use of draft replies makes sense. Choose the scope and evidence threshold before searching. Obtain a decision-maker’s approval if one is available; solo, record the choice as a fictional design decision. Completion check: geography, period, definitions and the minimum required evidence are verifiable, without suggesting that a real business has given approval.

Why can a reply with valid JSON still fail the evaluation from chapter 3?

A format check does not establish whether the content is supported by the source.
An unsupported promise or exceeding the task boundary can block the case, even when the format is technically correct.

4.2

Building a market map

Separate market definition, segments, size estimation, dynamics, and assumptions.

Create inclusion and exclusion rules and map customer segments, problems, providers, channels, substitutes, and rules. Use multiple estimates when hard data is lacking.

Label top-down, bottom-up, and proxy calculations with date, geography, and source quality. An exact number without a robust source is false precision.

A top-down estimate starts with a broader total and narrows it using justified proportions. A bottom-up estimate adds smaller building blocks, for example numbers of potential buyers × possible expenditure. Both produce a scenario. They do not prove purchase intent and must not be averaged to hide unexplained differences.

Market boundary
Segments
Substitutes
Estimate
Source quality
Sensitivity

Terms in plain language

Proxy: An indirect measure used to approximate something; check whether it is suitable for this purpose.

How you can use this

A Belgian service market is estimated bottom-up from numbers of enterprises and plausible annual expenditure.

Try this prompt

Create a market map for [market] in [region]. Define boundaries, segments, substitutes, drivers, and three estimates. Label source, year, and uncertainty.

Knowledge check

A fictional market map counts every business in a sector as a potential buyer. Whether they have the target problem and can be reached is unknown. How do you present that count?

As the reachable market, using sector membership as evidence that the businesses fit the chosen target group.As the expected customer base, with provisional average spending to calculate an initial revenue forecast.As a broad population boundary; need and reachability have not yet been established.

Your practical exercise

Calculate one segment top-down and bottom-up and explain the difference.

4.3

Competitor research with evidence

Compare claims only on the same definition, period, and source basis.

Separate direct competitors, alternatives, and doing nothing. Gather offer, target audience, price logic, channel, evidence, positioning, and visible limitations. Marketing language is not an independent performance measurement.

Retain the URL, publication date and date you checked the source. Mark price or capacity as unknown when it is not public; do not silently fill in a value.

Comparison set
Definitions
Source evidence
Date
Unknown
Alternatives

How you can use this

A matrix separates publicly stated prices, inferred prices and unavailable prices.

Try this prompt

Build a competitor matrix for [category]. Use the same criteria. For each cell, provide a source, date and status: confirmed, inferred or unknown. Include doing nothing as an alternative.

Your practical exercise

Check five competitor claims in the original source.

4.4

Structuring customer research

AI helps find patterns, but sampling and questioning determine what you learn.

Define target audience, recruitment, research question, and exclusions. Use interviews for depth, surveys for broader patterns, and behavioral data for actual actions. Do not ask leading questions.

Minimise data, arrange appropriate consent or another required legal basis and access, and connect themes to relevant evidence excerpts. Pseudonymisation may help but does not remove privacy requirements. A small qualitative sample is not necessarily representative. Use fictional interview excerpts for this exercise; process any real conversations only within the permitted arrangements.

Sample
Non-leading
Method
Evidence quote
Privacy
Limitation

How you can use this

Ask 'how did you solve this last time?' instead of 'would our AI solution be useful?'.

Try this prompt

Design customer research for [hypothesis] with target audience, recruitment, ten questions, privacy agreement, analysis method, and confirmation criteria.

Your practical exercise

Create three fictional interview excerpts with different experiences, code the statements and include counter-evidence. Real conversations are an optional next step with privacy arrangements agreed in advance.

4.5

Triangulating sources

More sources only help when origin and method really differ.

Compare source type, original publication, method, population, period, and interests. Ten websites repeating the same press release form one source chain.

In case of conflict, first look for differences in definition, period, or method. Make unresolved conflicts visible and state which additional evidence changes the decision.

Origin
Method
Period
Interest
Conflict
Additional evidence

Terms in plain language

Triangulation: Comparing a claim with evidence from sources with different origins or methods.

How you can use this

Two growth figures appear to use different regions and base terms.

Try this prompt

Triangulate [claims]. For each claim, map the source network with origin, method, period and interests. Explain conflicts and missing evidence.

Your practical exercise

Draw the source chains for three major claims.

4.6

From research to a decision memo

Separate evidence, interpretation, options, risk, and recommendation.

Start with the decision question and core answer. Show facts with sources, assumptions, alternatives, and what would change the conclusion. The recommendation follows from predetermined criteria.

Add counter-evidence, minority interpretation, and missing information. An authorized person makes the decision and archives source versions.

Decision question
Facts
Assumptions
Options
Counter-evidence
Recommendation

How you can use this

A market-entry memo recommends a limited pilot and identifies which evidence of demand is still too weak.

Try this prompt

Write a decision note from [research] with the core decision, evidence, assumptions, options, criteria, counter-evidence, risks, and reversible next step.

Your practical exercise

Check whether your decision memo examines decisive assumptions and alternatives. Ask a critical reader to review it if possible. Solo: put the memo aside and reconstruct its main conclusion from the source register in a separate round; also test the strongest alternative. Record only shortcomings you actually find. Completion check: the conclusion remains traceable and open questions are visible. Keep the source register and conflict memo with row H4; record their influence on the pilot choice and that solo review is self-assessment.

Worked example

Three excerpts are not automatically three pieces of evidence

Fictional practice material; incorrect answers have been created deliberately for this exercise.

The following source excerpts, organisations and dates are entirely invented for this exercise. They do not describe a real market. Atelier Noor is investigating whether customers need faster delivery before considering an additional service.

Input

Excerpt A, fictional price sheet from supplier Luma, version dated 1 March: “Standard delivery within five working days in the North zone.” Excerpt B, fictional promotional page from the same supplier, version dated 1 May: “Now delivering selected products within three working days in the Centre zone.” Excerpt C, fictional intermediary blog: “Luma delivers within three days”; the blog refers only to B. In addition, four of twelve customers in voluntary practice interviews name delivery time as a concern; three of those four ordered during the same busy week.

Deliberately flawed practice answer

“Three sources confirm that the market standard is delivery within three days. A third of all customers therefore want an express service.”

Check

A and B differ in date, zone and product coverage. They do not automatically establish a contradiction, but neither do they establish one general delivery time. C repeats B and is not independent confirmation. Working days are also different from calendar days. The twelve voluntary interviews have not been shown to represent all customers; the shared busy week may influence the answers. No excerpt describes the entire market.

Improved result

“This practice material shows different delivery promises for different situations. There is insufficient evidence of a market standard. Four of twelve interviewed customers mention delivery time, but participant selection and the order period limit the conclusion. Check the conditions for comparable products and postcodes, find an independent source and collect customer enquiries from several periods. For now, decide only to investigate further. Keep fact, interpretation and assumption separate in the decision memo.”

Try it yourself

A fourth fictional excerpt says: “For LAMP-2 in the North zone: three working days from 1 June”; it is a new approved price sheet from Luma. Which conclusion can you now add, and which broad conclusion remains unsupported?

View the model answer

You can document a new promise for LAMP-2 in the North zone from 1 June and check whether it replaces the older sheet. You still cannot say that all products, all zones or the whole market deliver within three working days.

Chapter assignment

Bring everything together

Submit a research brief, source register, competitor matrix, conflict note and decision document with a reversible next step.

My notes

Maximum 10,000 characters per note.

Progress and notes are stored only in this browser on this device. Do not enter sensitive data. Download your notes regularly. This course sets no automatic expiry date. You can delete the data through your browser’s site-data settings; export anything you wish to keep first. Browser settings or cleanup may erase it earlier. These local notes are not sent to Finaudax.