Market Validation from Oulu

Im being blunt: You are building a startup in Oulu but your market is not Oulu. It is too small. But Oulu is one of the best places in Europe to validate from: low burn rate, deep technical talent, institutional support for internationalization, and a tight community for fast feedback.

This playbook covers how to run a structured validation process using Jobs to Be Done (JTBD) and Outcome-Driven Innovation (ODI) — from Oulu, for international markets.

Oulu as Launchpad

Use Oulu for what it is good at:

Low cost base — run a 3-month validation sprint on a fraction of Helsinki or Stockholm budgets
Testing infrastructure — OuluHealth Labs, FabLab, 6G Flagship test networks. Build and test here, sell everywhere.
International connections — BusinessOulu, Business Finland, Nordic Innovation House all exist to connect you outward
Honest feedback — use local founders and mentors as sparring partners for your pitch and positioning

From the field: When I was building Willit, marketplace for local and natural produce, even that it the sellers and buyers are active for this trend right now, there actual demand is also at the other nordics

Step 1: Define Your Hypothesis for a Real Market

We believe that [customer segment] in [target market] has the problem of [pain point] and would pay [price] for a solution that [value proposition].

Target a market worth building for. "Nordic SaaS companies with 50-200 employees" is a hypothesis. "Companies in Oulu" is not.

Market	Why from Oulu
Finland	Beachhead — easy access, good public data, but too small alone
Nordics	Culturally similar, Business Finland support, strong trade links
EU	Large TAM, EU grants, university partnerships across Europe
US/Global	Nordic Innovation House soft-landing, Oulu's low cost base is an advantage

From the field: When I was building Willit I had clear that the initial one target market, like natural produce, farms, game / reindeer meat intrustry etc. is too small to be a viable target market. Just making calculations on how many people I could reach and looked in to other players, past failed startups. All helped on the journey.

Step 2: Start your Customer Discovery Interviews

Get into customer discovery interviews. For exampl use Oulu networks (StartupOulu, BusinessOulu, Business Finland, university contacts) for warm introductions to prospects.

Example 30-minute structure:

Context (5 min) — their day, their tools
Problem exploration (15 min) — last time they experienced the problem, what they did, what it cost
Solution reaction (5 min) — one-sentence concept, observe their gut reaction
Willingness to pay (5 min) — would they pay, how much, what would stop them

If fewer than 5 of 20 give strong signals ("When can I try this?"), rework the hypothesis.

I think that this actual human-to-human contact cannot be replaced by AI, but you can use GenAI to prepare for interviews and analyze them afterward. See the next steps.

From the field: I have been lucky that international customers have come to me for getting feedback and reach on some level. It helped that all websites have been made in multiple languages and the messaging was clear and compelling.

Step 3: Jobs to Be Done — Uncover the Real Demand

Standard interviews tell you what people say they want. JTBD tells you what they are actually trying to accomplish — the functional, emotional, and social jobs behind their behavior.

Job statement format:

When [situation], I want to [motivation], so I can [outcome].

The job is stable across geographies. Solutions change. If the same job appears in Helsinki, Stockholm, and Amsterdam, you have found something universal.

The Job Map

Map the full sequence your customer goes through:

Define — what triggers the need
Locate — finding inputs and resources
Prepare — setup before execution
Confirm — verifying readiness
Execute — the core action
Monitor — tracking progress
Modify — adjusting when things change
Conclude — finishing and what follows

At each step, look for friction — slow, unreliable, or frustrating. That is where your product wins.

GenAI Interview Packs

Use these prompts with Claude, ChatGPT, or any LLM. Copy, paste, fill in the brackets.

Pack 1: Prepare Your Interviews

I am an Oulu-based startup targeting [customer segment] in [target market].

My hypothesis: [your hypothesis]

1. Write 5 JTBD interview questions that uncover the functional job,
   emotional job, social job, and triggering context.
2. Note how questions might land differently by market
   (Nordic vs. US vs. Central European buyers).
3. Suggest 3 "switch interview" questions about the last time they
   switched solutions for this job.
4. Map the 8 job map steps for this specific job with one probe
   question each.

Pack 2: Analyze Each Interview

Here are my notes from a JTBD interview with [role, company, location]:

[paste notes]

1. Extract the core job statement:
   "When [situation], I want to [motivation], so I can [outcome]"
2. Identify emotional and social jobs.
3. Map to the 8-step job map. Flag highest-friction steps.
4. Write desired outcomes as ODI statements:
   "Minimize the [time/likelihood/effort] it takes to [outcome]"
5. Rate each: over-served, appropriately served, or under-served.
6. Flag geography-specific vs. universal findings.

Pack 3: Synthesize Across All Interviews

I completed [N] JTBD interviews targeting [segment] across [markets].

[paste all extracted jobs and outcomes]

1. Rank core jobs by frequency and emotional energy.
2. Segment interviewees by outcome priorities — do segments
   correlate with geography or are they market-wide?
3. Top 10 desired outcomes in ODI format, ranked.
4. Calculate Opportunity Scores (Importance + max(Importance - Satisfaction, 0)).
5. Flag over-served outcomes — do not compete there.
6. Recommend which job(s) to target for an Oulu-based startup
   building for international markets.

From the field: JTBD lists and with getAI has provided unique insights. As it could be little confusing to make large sets of interviews by hand.

Whats next? You have a prioritized list of outcomes that customers care about. Now connect those outcomes to the features you are building.

Step 4: Feature-to-Outcome Map (ODI)

Next I suggest that connect what you build to why customers care using Outcome-Driven Innovation. This is a survey-based method to prioritize features based on the outcomes customers want to achieve.

Desired Outcome Statements

[Direction] + [measure] + [object of control] + [context]

Example: "Minimize the time it takes to get a new customer to their first success metric after onboarding"

Can be difficult to get your head around. Here is genAi to the rescue again:

I am building [product] from Oulu for [segment] in [markets].
1. Write 10-15 desired outcome statements in ODI format based on JTBD interviews.
2. For each, suggest a survey question to measure importance and satisfaction.
3. Note any outcomes that seem to vary in importance or satisfaction by geography.

Score and Prioritize

For each outcome, survey your interview subjects:

Importance (1-10): how critical to getting the job done
Satisfaction (1-10): how well current solutions deliver

Opportunity Score = Importance + max(Importance - Satisfaction, 0)

Score	Meaning
15-20	Under-served — build here
10-14	Appropriately served — room to improve
0-9	Over-served — do not compete

Build the Map

Alright now what. We could rate each feature against each outcome (0 = no impact, 3 = direct). The features with the highest weighted scores across your top under-served outcomes become your MVP.

GenAI Prompt for the Map

Here is again prompt to use:

I am building [product] from Oulu for [segment] in [markets].

Desired outcomes ranked by Opportunity Score:
[paste outcomes with scores]

Candidate features:
[list features]

1. Create a Feature-to-Outcome matrix (0-3 scoring).
2. Calculate Feature Priority Score:
   sum(cell score × Opportunity Score) per feature.
3. Recommend MVP feature set for top 3-5 under-served outcomes.
4. Flag high-opportunity outcomes no feature addresses.
5. Note where satisfaction varies by geography — launch there first.

From the field: As a software engineer I love to build and still does. When doing my own startups focus changed from time to time, but it gets refined. What I learned that the features that is going to be builded has to be looked in to what is the income potential. It starts with gut feeling but the data helps refine it.

Now you have a prioritized list of features to build that are directly connected to the outcomes your customers care about. But how do you know if they will actually pay for it? That is the final step.

Step 5: Validate Willingness to Pay

Finally, gather evidence from your target market, not just local contacts. This is just a simple list that usually starts with:

Landing page - targeted ads to your market geography. 5%+ conversion is a positive signal.
LOIs - non-binding letters of intent from international prospects carry the most weight with Finnish VCs and Business Finland.
Pre-sales — if 3-5 companies pay before you build, you have validated demand.
Pilots — Finnish companies are culturally comfortable with pilots. Use them as first reference cases, then use those cases to open doors in the Nordics.
Public procurement — check Hilma for similar solutions being procured. Finnish public sector can be stable beachhead revenue.

From the field: This comes in my experience to trial and error, and main thing is to try and test hypothesis, but that every decision should have some concrete evidence to support it. I have builded large sets of data analytics that is build in to the systems internally. I dont trust 3rd party services so much