Step 2. Map the decisions that matter
Great analytics informs important decisions. To know which decisions matter, you need to understand how the business actually operates day to day.
The best way to do this is to map the journey a customer takes with the business from first finding out about it through to eventually leaving. Every part of the business exists to serve that journey in some way, which makes it the right starting point.
Once the journey is mapped, you annotate it with layers of context pulled directly from the BMC you completed in Step 1. Each layer makes the picture more complete, and together they surface the questions worth asking.
The five annotation layers
Work through the layers in this order, because each one builds on the previous:
- Customer journey. The stages a customer moves through from first contact to exit. This is the foundation.
- Key activities. For each stage, what does the business actually do? What decisions or capabilities make the stage work?
- Partners and resources. Who and what does the business depend on at each stage? This surfaces where you're exposed if something breaks.
- Costs and revenue. Is each stage a cost, a revenue event, or both? Where does money come in and where does it go out?
- Segments and value. Which customer segments move through each stage, and what do they get out of it?
The questions worth asking tend to hide at the intersections of these layers. A question that only becomes visible when you see that a particular stage depends on an external partner, costs money per transaction, and serves two segments differently is a much more valuable question than anything you'd find on the surface.
We'll walk through both of our running examples below.
Example 1 — B2B SaaS
Customer journey
| Stage | Description |
|---|---|
| Marketing | Paid and organic activity drives awareness and brings potential customers to the product. |
| Sales | For larger accounts, a sales rep gets involved: outreach, demos, proposals, and negotiation before the account is won. |
| Signup | A visitor starts a free trial. No payment required upfront. |
| Onboarding | The new user sets up the product and gets it working for their team. |
| Activation | The user reaches the first moment where the product delivers real value (for example, completing their first project or running their first report). |
| Ongoing usage | The team uses the product regularly as part of how they work. |
| Expansion | The account grows: more users, higher plan, or additional features. |
| Renewal | The subscription comes up for renewal and a decision is made to stay or leave. |
| Churn | The account cancels or stops paying. |
| Win-back | A former customer is re-engaged and returns. |
Notice that Ongoing usage can branch in multiple directions: strong engagement leads to Expansion and Renewal; weak engagement leads to Churn.
Annotated process map
| Stage | Key Activities | Partners & Resources | Costs & Revenue | Segments & Value |
|---|---|---|---|---|
| Marketing | Campaign creation, channel selection, budget management | Ad platforms, content partners; marketing team, analytics tools | Cost: ad spend, content production | Reaches all segments; value = awareness and intent |
| Sales | Outreach, demos, proposals, contract negotiation | CRM; sales team | Cost: sales salaries and commissions | Larger accounts only; value = a buying decision that self-serve wouldn't achieve |
| Signup | Trial sign-up flow, welcome email sequence | Product, email automation | Cost: hosting and support | Self-serve by all segments; value = low-friction entry |
| Onboarding | In-app setup guides, onboarding emails, setup calls for larger accounts | Customer success team, onboarding tools | Cost: staff time for larger accounts | Smaller accounts self-serve; larger accounts get hands-on help |
| Activation | Feature adoption tracking, nudges for users who haven't hit the activation milestone | Product analytics | Cost: engineering and product time | All segments; value = first real outcome from the product |
| Ongoing usage | Product improvements, support, reliability | Cloud infrastructure, integration partners; engineering and support teams | Cost: infrastructure, support; Revenue: monthly subscription fees | Power users drive retention; low-engagement users are the churn risk |
| Expansion | Upsell conversations, usage-triggered prompts | Account managers, CRM | Cost: sales time; Revenue: additional seats or plan upgrades | Mid and large accounts; value = more capability as the team grows |
| Renewal | Renewal conversations, business reviews | Account managers | Cost: customer success time; Revenue: renewed annual subscription | All segments; large accounts get proactive outreach |
| Churn | Exit surveys, offboarding | CRM | Cost: lost revenue | Primarily smaller accounts; value = lessons for improving retention |
| Win-back | Re-engagement campaigns, targeted offers | Marketing tools, CRM | Cost: campaign spend; Revenue: recovered subscriptions | Former customers who left for fixable reasons |
Example 2 — Non-bank lender
Customer journey
| Stage | Description |
|---|---|
| Marketing | Campaigns generate awareness among borrowers and brokers through digital channels and rate comparison sites. |
| Referral | A broker or existing customer refers a new borrower to the lender. For most non-bank lenders this is the dominant source of new business. |
| Application | A borrower applies directly through the lender's website or submits an application through a broker. |
| Pre-approval | An initial assessment is done using publicly available credit data and the borrower's declared information. The borrower gets an early indication of whether they'll be approved. |
| Document collection | The borrower provides supporting documents: proof of identity, income, expenses and assets. |
| Approval | A full assessment is done against the verified documents. A formal decision is made, conditions are set, and the interest rate is confirmed. |
| Settlement and funding | Loan documents are signed, the security (the property) is registered, and the money is disbursed. |
| Servicing | The borrower makes regular repayments. The lender handles payment processing, customer enquiries and account changes. |
| Hardship | A borrower in financial difficulty is assessed and placed on a modified repayment arrangement. |
| Arrears | A borrower who has missed payments is contacted and offered a path back to good standing. |
| Collections | Formal recovery action for borrowers who haven't responded to earlier intervention. |
| Discharge or refinance | The loan is paid off and the account is closed, or the borrower refinances (either staying on new terms or moving to a competitor). |
Note the branches after Servicing: most borrowers move toward Discharge or Refinance, but some move into Hardship or Arrears first.
Annotated process map
| Stage | Key Activities | Partners & Resources | Costs & Revenue | Segments & Value |
|---|---|---|---|---|
| Marketing | Campaign management, broker relationship activity, listings on comparison sites | Ad platforms, comparison sites; marketing team, CRM | Cost: ad spend, broker events | Reaches all borrower segments and brokers; value = awareness |
| Referral | Managing broker relationships, accrediting new brokers, tracking referral volume and quality | Broker networks and aggregators; broker portal, CRM | Cost: broker commissions; Revenue: none yet, but the primary driver of new loans | Broker-referred borrowers are the majority; existing customers referring friends and family are a smaller but high-trust source |
| Application | Managing the online application portal, supporting brokers through submission | Broker networks; online portal, broker portal | Cost: portal maintenance, broker support | Broker channel dominates for most segments; direct channel attracts self-employed and digitally savvy borrowers |
| Pre-approval | Running an initial credit assessment, issuing an indicative decision | Credit bureaus; credit assessment system | Cost: credit check fees | All borrower types; value = fast early answer, especially important for time-sensitive purchases |
| Document collection | Chasing outstanding documents, verifying what's received | Identity and income verification providers, property valuers; document management system | Cost: staff time, verification fees | Self-employed borrowers with complex income take longer and create more work here |
| Approval | Full credit assessment, setting pricing, issuing formal approval | Property valuers, mortgage insurance providers; credit assessment system | Cost: staff time, valuation and insurance fees | Near-prime borrowers require more manual assessment |
| Settlement and funding | Preparing loan documents, coordinating settlement, drawing down funds | Lawyers, conveyancers, settlement platform providers; settlement system, loan system | Cost: legal and settlement fees; Revenue: establishment fee collected | All segments; value = loan formally in place, money disbursed |
| Servicing | Processing repayments, handling enquiries, managing rate changes | Payment processors; loan servicing system | Cost: servicing staff; Revenue: ongoing interest margin | Long-term relationship with all borrowers; value = smooth, low-friction experience |
| Hardship | Assessing the borrower's situation, negotiating a modified arrangement | Financial counselling services; case management system | Cost: staff time, reduced margin on the arrangement | Borrowers in financial difficulty; value = a path through a hard period |
| Arrears | Early contact, repayment plan discussions | Collections contact system | Cost: staff time | Borrowers who have missed payments; value = getting back on track before things escalate |
| Collections | Formal recovery process | Debt collection agencies, lawyers; collections tracking system | Cost: agency and legal fees; Revenue: partial recovery | Borrowers who have not responded to earlier contact |
| Discharge or refinance | Processing the payout, releasing the security, making a retention offer | Servicing system | Cost: staff time; Revenue: ends at discharge | All segments; refinance is a retention opportunity |
Finding the questions that matter
The annotated process map gives you the raw material. Now you need to turn it into questions worth answering.
Step 1: Come prepared, then interview the operators. Before you talk to anyone, use your annotated process map to draft your own version of the questions for each stage. What would you want to know if you were running that part of the business? Write them down. They will be incomplete and some will be wrong — that is the point.
Then sit down with the person who runs each stage: the head of marketing, the operations lead, the sales manager. Don't send a survey. Have the conversation. Show them your draft questions and ask them to tell you what you have missed, what is wrong, and what actually matters to them. A stakeholder who is correcting your thinking will give you far more than one who is answering open questions from a blank page.
Use these prompts to push past the surface answers:
- What decisions do you make repeatedly that you feel you are making on gut instinct?
- Where do you feel like things are going wrong but you cannot quite put your finger on why?
- What would you want to know at the start of each week that you do not currently have?
- If you had a data analyst sitting next to you full time, what would you ask them first?
Step 2: Separate the two types of question. The questions you collect will fall into two groups, and you need both.
The first group is about understanding performance: what is happening, is it working, where are things going wrong? These are descriptive and diagnostic. Essential, but most businesses have more of these than they have answers.
The second group is about informing the next decision: what will happen, who should we prioritise, where should we act? These are predictive and prescriptive. Most businesses are underinvested here, and this is where analytics creates its most distinctive value.
After someone gives you a performance question, follow up with: "If you knew the answer to that, what would you do differently?" That almost always surfaces the more valuable predictive version underneath.
Step 3: Look for the intersections. The best questions hide where two layers of your annotated map overlap: where a cost driver affects one segment but not another, where a partner's reliability creates a revenue risk, where a stage serves two segments with very different needs. These intersections are where analytics finds things that gut instinct misses.
Example questions — B2B SaaS
| Stage | Understanding performance | Informing the next decision |
|---|---|---|
| Marketing | Which campaigns are driving signups, and what is the cost per trial? | Which campaigns are most likely to bring in accounts that convert and stay? Where should we shift budget? |
| Sales | What is our win rate by deal size and segment? Where are we losing? | Which prospects in the pipeline are most likely to close, and which should the team prioritise this week? |
| Signup | Where are people dropping off in the signup flow? | Which visitors are most likely to complete signup if we intervene? |
| Onboarding | What proportion of new accounts complete onboarding, and how long does it take? | Which new accounts are struggling and need intervention before they disengage? |
| Activation | What is our time to first value, and how does it vary by segment? | Which users are at risk of never activating, and what nudge is most likely to help? |
| Ongoing usage | Which features are being used and which are being ignored? | Which product investments would have the biggest impact on retention? Which disengaged users are about to churn? |
| Expansion | Which accounts have grown usage without upgrading their plan? | Which accounts are ready to expand, and what is the right offer and timing? |
| Renewal | What is our renewal rate by segment and cohort? | Which accounts are at risk of not renewing, and what should we do about each one? |
| Churn | Why are customers leaving, and does the reason vary by segment? | Which active accounts are most likely to churn in the next 90 days? |
| Win-back | What is our win-back rate, and which types of former customers return? | Which churned accounts are worth pursuing, and what offer has the best chance of working? |
Example questions — Non-bank lender
| Stage | Understanding performance | Informing the next decision |
|---|---|---|
| Marketing | Which campaigns are generating applications that go on to settle? What is the cost per settled loan? | Where should we shift budget to get more of the right applications? |
| Referral | Which brokers are sending the most volume and the best quality applications? | Which brokers should we invest more in, and which are costing more than they are worth? |
| Application | What is application volume by channel and how is the mix trending? | Are there sources of applications we are underinvesting in? |
| Pre-approval | What proportion of pre-approvals proceed to full approval, and where do they fall over? | Which applications are likely to fail at approval so we can flag them early? |
| Document collection | How long is document collection taking, and where are the delays? | Which applications are going to miss settlement targets unless we intervene now? |
| Approval | What is our approval rate, and how does credit performance compare to what we expected? | Are there applications in the queue that we should prioritise or escalate? |
| Settlement and funding | What proportion of settlements are on time, and where do delays come from? | Which loans in the pipeline are at risk of not settling on time? |
| Servicing | How is the book performing by cohort, product, and channel? | Which borrowers are showing early signs of stress before they miss a payment? |
| Hardship | What proportion of hardship arrangements result in the borrower returning to normal repayments? | Which borrowers are at risk of entering hardship in the next three months? |
| Arrears | Which contact approaches are most effective at resolving arrears? | Which borrowers in arrears are most likely to resolve without escalation, and which need immediate action? |
| Discharge or refinance | How many customers are leaving at refinance and what is the typical timing? | Which borrowers are most likely to refinance away in the next six months, and is there an offer that would retain them? |
In Step 4, you will take this list of questions and turn each one into a defined analytics use case — with a named stakeholder, a pattern, and an example approach.