Mining The Problem Space

    Stefan Kalpachev

    Stefan Kalpachev

    Founder & CEO, Content RevOps

    •
    April 13, 2026
    •
    5 min read

    Once we have a likely ICP, the next job is to understand their problem space properly.

    By “problem space”, we mean the real-world context around the buyer’s problem:

    • what they are actually trying to achieve

    • what gets in the way

    • what pressures they are under

    • how they describe the problem in their own words

    • what success looks like to them

    • where they look for ideas, answers, and reassurance

    This is where we move beyond a surface-level persona.

    A persona tells you who the audience is.
    The problem space tells you what is going on in their world.

    That matters because a strong strategy comes from understanding the buyer’s reality, not just their job title.

    Our rule on AI

    We do not let AI invent the audience for us.

    AI can help us organise, summarise, and spot patterns.
    But the useful part has to come from real evidence first.

    That means we start with actual inputs:

    • customer conversations

    • sales conversations

    • delivery or account insights

    • feedback data

    • signs of real audience behaviour online

    AI comes in later, once there is something real to analyse.


    Where we get the evidence from

    We collect evidence in a rough order of trust.

    1. Customer interviews

    If possible, we speak to real customers first.

    We want to understand:

    • what they were trying to solve

    • what made the decision difficult

    • what alternatives they considered

    • what changed after they bought

    This gives us the clearest view of the real job behind the purchase.

    2. Sales conversations

    Next, we speak to the people closest to the buying process.

    Usually that means the founder, sales lead, or anyone handling sales calls.

    They can usually tell us:

    • what objections come up repeatedly

    • what buyers worry about

    • what signals real intent

    • what makes deals move or stall

    3. Customer success or account management

    These conversations help us see what customers thought they needed versus what they actually needed once delivery started.

    That gap is often useful.

    4. Existing feedback and notes

    Then we go through any existing internal material:

    • surveys

    • feedback forms

    • CRM notes

    • onboarding notes

    • support tickets

    • call summaries

    This helps us move from isolated opinions to recurring patterns.

    5. Online audience behaviour

    After that, we look at the audience in the wild.

    We want to see:

    • where they spend attention

    • what they complain about publicly

    • what they engage with

    • what language they use when they are not being interviewed

    This usually includes places like:

    • LinkedIn

    • Reddit

    • niche publications

    • newsletters

    • podcasts

    • relevant job descriptions, where useful

    How we work through it

    Step 1: Build one raw evidence document

    We keep all the inputs in one place.

    Usually that means one working doc or folder with sections for:

    • customer interviews

    • sales interviews

    • customer success or account notes

    • survey or feedback data

    • online audience research

    • useful market material

    The important part is simple:
    raw evidence stays separate from interpretation.

    We do not want polished summaries too early.

    Step 2: Identify the jobs to be done

    Next, we ask what this audience is really hiring the product or service to help them do.

    We look for questions like:

    • What are they trying to achieve?

    • What does “done well” look like for them?

    • What are they trying to avoid?

    • What happens if they do nothing?

    We try not to stop at the obvious functional answer.

    For example, the job is rarely just:
    “buy software”
    or
    “hire an agency”

    Usually it goes deeper:

    • reduce risk

    • save time

    • avoid internal friction

    • protect reputation

    • hit a target they are accountable for

    • feel more confident in a difficult role

    That deeper layer is often where the best strategy comes from.

    Step 3: Identify frustrations and pressure points

    Then we look at what makes this problem difficult.

    We pull out things like:

    • recurring complaints

    • repeated blockers

    • anxieties

    • fears

    • frustrations with current options

    • situations where they feel exposed or blamed

    This matters because buyers do not act on logic alone.

    A lot of buying urgency comes from pressure, uncertainty, and frustration.

    Step 4: Understand how they see the world

    We also want a clearer picture of how this audience thinks.

    That includes:

    • how they describe themselves

    • what they take pride in

    • what they see as good work

    • what they distrust

    • what kind of information they respond well to

    This helps us avoid producing content that is technically relevant but culturally wrong.

    Step 5: Capture their actual language

    We always create a section for verbatim language.

    This includes exact phrases around:

    • pains

    • goals

    • objections

    • desired outcomes

    • bad alternatives

    • how they describe their own role

    This is one of the most useful parts of the process.

    It gives us language we can later use in:

    • messaging

    • positioning

    • copy

    • content angles

    • sales material

    Step 6: Use AI to spot patterns

    Only once we have enough raw material do we use AI to help analyse it.

    At that point, AI is useful for things like:

    • grouping recurring jobs to be done

    • spotting repeated frustrations

    • separating practical pains from emotional ones

    • summarising media habits

    • pulling out repeated phrases

    • highlighting early content opportunities

    The key is that we use it for pattern analysis, not for making things up.

    We want the raw truth first, and the neat summary second.

    What we end up with

    By the end of this stage, we should have a much clearer picture of the audience’s reality.

    That usually includes:

    • the main job they are trying to get done

    • secondary jobs around it

    • common frustrations and emotional pressures

    • how they think about success

    • how they describe the problem

    • where they spend attention

    • what they trust

    • which themes keep coming up again and again

    At that point, we are no longer working from a vague persona.

    We are working from a grounded view of:

    • the buyer’s world

    • the buyer’s language

    • the buyer’s real decision context

    That gives the rest of the strategy something solid to build on.