Full name

Use the same name all week — it is how your submissions are matched. Each export names its own file from it, for example 1-muchson-day1-l1-diagnostic.

Route 1 · Level 1 · Knowledge

Read what the file records

Materi A · Level 1 · Knowledge · 60 min

What a customer file records, and what it does not

Eight short cards (58 min of reading and exploring, the rest is discussion). Each carries a named framework, a real figure or a real case, and each diagram is a tool the task below reuses.

Why retention economics differ in IT project business

6 min

In project business, revenue stops at unless something is designed to continue it.

In plain words

A project pays you while you build it. The day the customer signs it off (go-live), the invoices stop. If nothing is planned for after that, the customer drifts away and the money goes with them. Retention means planning what comes after go-live before you get there.

Why it matters

At TechSolutions, 70% of customers buy once and never come back. Keeping even a few more of them is worth a lot, because a customer who stays keeps paying year after year.

How to read the picture below

The bars show one project's revenue: high during delivery, then zero after go-live. Drag the slider for r, the share of customers who stay each year, and watch (what one customer is worth over time). Moving r from 0.80 to 0.85 already adds about 27%: a small change in who stays makes a big change in value.

Revenue cliff

Retention rate r: 0.80

m = yearly margin €30,000 · i = discount rate 10% (Case assumption). Drag r, the retention rate (the share of customers who stay each year). Not graded.

CLV = m · r / (1 + i − r)

€80,000

reference point: r = 0.80

What this showsThis is the reference point (r = 0.80, CLV = €80,000). Drag the slider left or right to see how much a small change in retention rate moves this figure.

The shape is illustrative. The readout uses the Gupta & Lehmann formula with inputs.

Project business sells outcomes once: fixed scope, , acceptance (), then revenue drops unless a exists. Case company TechSolutions GmbH: 70% of customers are .
The classic argument: Reichheld & Sasser (1990) report that cutting by 5% raised profits by 85% in one bank's branch system, 50% in an insurance brokerage, 30% in an auto-service chain; MBNA halved a 10% defection rate and profits rose 125%.
Simple (CLV) model, Gupta & Lehmann (2003): CLV = m · r / (1 + i − r) (constant m, r, i, infinite horizon). In plain words: the total profit one customer brings you, from the yearly profit m, how many customers stay each year (r), and how much later money is worth less (i). At r = 0.80, CLV = €80,000; at r = 0.85, €102,000 (+27.5%).

Read it carefully

The popular slogan “+5% retention = +25–95% profit” is a distortion of those company-specific figures from consumer-service industries, and the calculation was later disputed (Carroll & Reichheld 1992). Treat it as direction, not as a forecast for your firm.

How to decide when this comes up in the task

A project ends its revenue at acceptance (). Retention has to be designed in; it does not happen by default.
Treat the “+5% retention = +25–95% profit” slogan as direction, never as a forecast for your firm.

Satisfaction, loyalty, retention: three different things

9 min

Satisfaction is an evaluation, loyalty is an attitude with behaviour, retention is an observable fact, and they can diverge.

In plain words

Three words that sound alike but mean different things. Satisfaction asks “was I happy with the result?”. Loyalty asks “do I like you and keep coming back?”. Retention is simply “did they buy again?”, which you can see in the records. A customer can be happy and still leave, or unhappy and still stay.

Why it matters

Mix them up and you trust the wrong signal. A survey score of 4 out of 5 feels safe, but it does not tell you the customer will order again.

How to read the picture below

The square has two axes: how much the customer likes you (left to right) and how often they buy again (bottom to top). Click each box for an IT example. Then pick a scenario and watch customer Z move: a price cut or a lost contact can pull liking and buying apart.

Loyalty quadrant

What this showsLoyalty. B2B IT example: Recommends the supplier and re-orders.

Scenario for customer Z Exploratory · not graded

What this showsChoose a scenario to move the marker.

Select a quadrant for a B2B IT example, then run a scenario.

Three constructs
Construct	What it is	Typical measure	B2B IT divergence example
Satisfaction	Judgement of outcome vs expectation	(1–5), close-out survey	4/5 and no re-order.
Loyalty	plus (Dick & Basu 1994); Oliver (1999) describes four stages: cognitive → affective → conative → action loyalty	(Reichheld 2003), intention,	Recommends the supplier, yet a procurement policy forces three offers.
Retention	Observable continuation	, within 24 months, / for	Stays only because exit is expensive.

Jones & Sasser (1995): Xerox found “totally satisfied” (5) customers were six times more likely to repurchase within 18 months than “satisfied” (4). Typology: (satisfied + loyal), (satisfied, not loyal), (dissatisfied, stay because of barriers), .
caveat: it records stated intent, and in project business the respondent (often the head of IT) is rarely the .

Metric definitions

= (starting − − contraction) ÷ starting recurring revenue. adds expansion. Project business often has no recurring revenue, so use and .

How to decide when this comes up in the task

A high satisfaction score does not prove the customer will re-order: a 4 out of 5 is not a safe score.
Retention is what the records show (renewal, repeat orders). Do not read attitude into it: a customer can stay only because exit is expensive.
In project business the survey respondent (often the head of IT) is rarely the , so a stated intention is weak evidence.
Read each construct from the record that measures it: satisfaction from a survey or rating; from , stated intention or ; retention from orders and renewals. If no record in the file measures a construct, it is “not recorded”, and the loyalty quadrant cannot be placed on that axis.
Place a customer on the quadrant from the two axes you have read, never by picking the quadrant first. An axis with no record stays open: the answer is then a band across two quadrants, not a point.

Three types of retention

9 min

Customers are held by three different ropes, and each rope breaks under a different trigger.

In plain words

Picture a customer tied to you by up to three ropes. Emotional: they trust someone on your side. Economic: leaving would cost them too much effort or money. Contractual: a signed contract keeps them. Most customers hang on a mix, with some ropes thick and others thin.

Why it matters

Each rope snaps for a different reason, so each needs a different fix. A longer contract does nothing if what really held the customer was one trusted contact who has just left.

How to read the picture below

Switch between Customer P and Customer Q and compare how thick each rope is. Then click a trigger (a price cut, a contact leaving, a contract ending) to fray a rope, and see which ropes still hold.

Three ropes

Customer P: three-year managed-service contract, weak relationship.

What this showsCustomer P is held mainly by the contract (strength 3 of 3) and the switching cost (strength 2); the relationship itself is weak (strength 1). Take the contract away and little remains. The same fix would not work on both: a longer contract protects P but does nothing for Q, whose rope is emotional, not contractual.

Trigger Exploratory · not graded

What this showsStill holding firmly: economic, contractual. Each rope breaks under a different trigger, and a trigger on a rope that was already thin changes little.

Two demo customers, three triggers. Select a trigger to fray its rope.

Three types of retention
Type	What holds the customer	Anchor	B2B IT example	Breaks when	Data signal
Emotional	Trust and affinity	; trust = ability, benevolence, integrity (Mayer, Davis & Schoorman 1995); commitment-trust theory (Morgan & Hunt 1994)	A named contact the customer relies on	The contact leaves, or silence follows	Contact cadence, sponsor continuity
Economic	Rational calculation	: leaving costs more than staying (Gustafsson, Johnson & Roos 2005)	Integration, retraining, re-certification effort	A rival's discount exceeds the	gap vs switching cost
Contractual	Formal obligation	(Rahmenvertrag), term, (Kündigungsfrist)	A 3-year service contract with 6-month notice	The term ends, so renewal becomes a new procurement event	Contract end date, notice window

Gustafsson et al. (2005) separate affective from and study triggers such as price increases and service failures.

How to decide when this comes up in the task

Name the rope before the fix: contact that has stopped is an emotional-rope problem; a rival's discount is an economic one; a term that ends is a contractual one.
Retention that rests only on economic or contractual ropes is the “hostage / spurious” case from A2, not loyalty.
A data signal points at a rope: contact cadence and sponsor continuity for emotional; the gap versus for economic; end date and notice window for contractual.
The economic rope needs both halves of the comparison: a rival's price gap and the customer's switching cost. A file that records only the price gap does not show the economic rope; it is “not recorded”.
No service or on file means the contractual rope is absent; a tender at the end is a new procurement event, not a renewal.
Emotional rope: count what still works. If some contact, owner or channel is left, the rope is weak; if none is left, it is absent.

Switching costs in project business

9 min

are highest while you deliver and tend to fall after handover unless value is renewed.

In plain words

Switching cost is everything a customer loses by changing supplier: time and effort (learning new tools, running a new tender), money (exit fees, lost discounts) and people (a contact they trust). It is highest while you are delivering the project and slowly shrinks after handover.

Why it matters

Once switching cost has dropped, a competitor's offer suddenly looks attractive. If you do not keep adding value after , the customer can leave easily, and often quickly.

How to read the picture below

Turn the two lines on and off. Both rise the same way during delivery. After go-live the red line falls (no new value), while the teal line stays high (value keeps being renewed). The curve is a thinking model, not measured data.

Switching-cost curve

What this showsBoth lines start together — switching cost peaks the same way during delivery no matter what happens later. They only split apart after go-live: renewed value holds the line up, its absence lets it decay.

Switch the two lines on and off.

Burnham, Frels & Mahajan (2003) typology: (economic risk, evaluation, learning, setup), financial (benefit loss, monetary loss), (personal and brand relationship loss).

IT mapping
In an IT project	Type
Learning a new operator's tooling	· learning
Environment discovery and	Procedural · setup
Re-running a tender	Procedural · evaluation
Uncertainty about a new supplier's performance	Procedural · economic risk
Exit fees or lost discounts	Financial
Losing a trusted contact

Working model — reasoning, not a measured law

In project business, peaks during delivery (fresh integration, undocumented know-how) and decays after handover when documentation moves to the client and tacit knowledge fades. Jones & Sasser (1995): barriers produce , and when the barrier falls can be fast.

How to decide when this comes up in the task

Separate the three types: (time and effort), financial (money and benefits lost), (people and brand lost). A rival's discount has to beat all three.
The decay after handover is a working model, not a measured law: use it to ask what renews value, not to predict a date.

The customer journey in B2B IT

9 min

The next purchase is shortlisted in the stretch of the journey that most project firms do not staff.

In plain words

The customer journey is the whole path a customer walks with you, from “we need something” to “do we buy again?”. In an IT project it has seven stages: tender, offer, contract, delivery, , operating, and renewal.

Why it matters

Most project firms stop paying attention after go-live, in the “Operate & review” stage, where nobody is really in charge. Yet that is exactly when the customer quietly decides who to invite next time. Buyers spend only about 17% of their buying time talking to suppliers; the rest happens without you.

How to read the picture below

Click a stage on the map to read what happens there and who on your side owns it. Look for the stage marked as often unassigned: that gap is the point of this card.

B2B journey

What this showsSelect a stage (tap, click or press Enter) to read its caption, or a record chip below the axis for when it happened.

Brand-owned
Partner-owned
Customer-owned
Social / external

Who owns each stage is a practitioner observation, not a statistic.

The same JourneyMap component is used in Task 1 with the Kessler file. Select a stage for its caption.

Lemon & Verhoef (2016): the journey runs across pre-purchase, purchase and post-purchase; are brand-owned, partner-owned, customer-owned, or social/external.
B2B IT project stages (used in Task 1): Need & tender () → Offer & negotiation → Contract → Delivery → & → Operate & review → Renewal / re-tender.
Gartner (2017 Digital B2B Buyer Survey, n = 750): buyers spend about 17% of buying time meeting suppliers; the rest is independent research and internal alignment. What happens between your touchpoints is where the next shortlist forms.

How to decide when this comes up in the task

Read a journey by stage and by owner: for each stage ask who on your side owns it, and what record shows contact in it.
A gap between two is not empty for the buyer: buyers spend about 17% of buying time meeting suppliers, and the rest on independent research and internal alignment.

Observation vs interpretation, and how to sort (MECE)

9 min

Evidence is a record that survives without the speaker. Everything else is interpretation.

In plain words

An observation is something written in a record that anyone can check, like “invoice 2231 was paid 41 days after issue”. An interpretation is what someone thinks it means, like “they pay late because they are in trouble”. The first is evidence; the second is a guess until a record backs it up.

Why it matters

In Task 1 you sort a customer file. Treat guesses as facts and your diagnosis points at the wrong problem. is just a way of sorting into boxes that do not overlap and together cover everything.

How to read the picture below

Click the rungs of the ladder from bottom to top. Each step up adds a little more of your own meaning, until you end up acting on a belief. Then try the four Alpenwerk sentences further down: tag each as observation or interpretation, and press Reveal to compare.

Ladder of inference

Rung 1 of 7

Observable data

What a record shows, without the speaker.

Example (Alpenwerk GmbH, illustrative): “Invoice 2231 was paid 41 days after issue (terms: 30 days).”

What this showsOnly this rung survives without the speaker. Everything above it is interpretation — a claim that needs its own evidence before you can rely on it.

Select a rung. The bottom rung is the only one that is a record.

(Argyris 1990, popularised by Senge 1990): observable data → selected data → added meanings → assumptions → conclusions → beliefs → actions.
Three tests (form only): Who recorded it, when, in which document? Would two readers of the same record agree? Does the sentence name a cause the record itself does not show?
(Minto 2009): categories that do not overlap and together cover the field.

Sorting key for Task 1 · five bins

Price: Money the customer pays or compares.
Trust: Evidence of delivery reliability or conduct (deadlines met, promises kept, incidents handled): ability, benevolence, integrity. (Mayer et al. 1995)
Benefit: The outcome the customer receives and rates (needs met, value in use). (Anderson, Narus & van Rossum 2006)
Relationship: Continuity of contact between the two organisations: people, cadence, ownership.
Interpretation — not evidence: A claim the file does not show. The note may exist; the cause it names is not recorded.

Try it — Alpenwerk GmbH Exploratory · not graded · not exported

“Invoice 2231 was paid 41 days after issue (terms: 30 days).”
“Alpenwerk pays late because it is in financial trouble.”
“The project log records that the go-live date moved by two weeks.”
“Alpenwerk's IT lead has lost interest in the project.”

How to decide when this comes up in the task

Test a sentence with the three tests: who recorded it, when, in which document? Would two readers of the same record agree? Does it name a cause the record itself does not show?
A personal comment (a note, a handover remark) is an interpretation unless a record in the file shows the cause it names. It may exist as a note; it is not evidence.
Sort by what the record is about: money → Price; deadlines kept and incidents handled → Trust; the outcome the customer rates → Benefit; continuity of people, cadence and ownership → Relationship.
A record can be a fact about a rival's price and still not be the weak point on your side: name the category holding the most observed evidence of a shortfall on TechSolutions' side, and cite records, not remarks.

Compliance corner: retention outreach in Germany/EU

5 min

Retention needs contact, and contact needs a legal basis.

In plain words

To keep customers you have to stay in touch, but in Germany you cannot simply send marketing emails, not even to business addresses. Usually you need the person's consent first. There is one exception for existing customers, and it only applies when all four of its conditions are met.

Why it matters

A retention campaign that breaks the law creates a bigger problem than the one it solves. Service messages, such as a security update, are fine; advertising is where the rules apply.

How to read the picture below

Answer the four yes/no questions. Every “yes” takes you one step down; a single “no” means you need consent. Press Start again to try another path. It is a rule of thumb, not legal advice.

May we send this?

Question 1 of 4

Is the message advertising?

A walk-through of the rule of thumb, not legal advice.

Answer the four questions to walk the path.

Business contact data (name@company) is personal data under . Art. 6(1)(f) (Recital 47 recognises direct marketing as a possible legitimate interest); Art. 21(2)–(3) right to object to direct marketing; Art. 5(1)(c) and (e) and . Stale contact data is both a retention problem and a compliance problem.
Email advertising in Germany (§ 7 ) requires prior consent. B2B is not exempt. The existing-customer exception § 7(3) UWG needs all four: address obtained in connection with a sale; used for own similar products/services; no objection; clear information at collection and at every use.
Working rule: keep service messages (, security advisories) separate from advertising.

Rules of thumb, not legal advice

Involve the (data protection officer).

How to decide when this comes up in the task

Advertising email to a business address needs prior consent in Germany. B2B is not exempt.
The existing-customer exception, § 7(3) , needs all four conditions. Miss one and consent is needed.
Keep service messages separate from advertising.

Field method for Task 1

2 min

Four steps, fifteen minutes: Read → Tag → Sort → Name.

In plain words

A short recipe for Task 1. Read every record in the file, tag each one as observation or interpretation, sort them into the five bins, and finally name the one area where the evidence shows TechSolutions fell short the most.

Why it matters

You have 15 minutes. Following the steps in order stops you from jumping to a conclusion before you have read everything.

Time box: 15 minutes. Task 1 follows directly.

How to decide when this comes up in the task

Read every record, tag each as observation or interpretation, sort into the five bins, then name the category with the most observed evidence of a shortfall and cite two records.

References · Materi A (19)

Anderson, J. C., Narus, J. A., & van Rossum, W. (2006). Customer value propositions in business markets. Harvard Business Review, 84(3), 90–99.
Argyris, C. (1990). Overcoming Organizational Defenses. Allyn & Bacon.
Senge, P. (1990). The Fifth Discipline. Doubleday.
Burnham, T. A., Frels, J. K., & Mahajan, V. (2003). Consumer switching costs: A typology, antecedents, and consequences. Journal of the Academy of Marketing Science, 31(2), 109–126.
Carroll, P., & Reichheld, F. F. (1992). The fallacy of customer retention. Journal of Retail Banking.
Dick, A. S., & Basu, K. (1994). Customer loyalty: Toward an integrated conceptual framework. Journal of the Academy of Marketing Science, 22(2), 99–113.
Gartner (2017). Digital B2B Buyer Survey (n = 750), as reported in Gartner buyer-enablement materials.
Gupta, S., & Lehmann, D. R. (2003). Customers as assets. Journal of Interactive Marketing, 17(1), 9–24.
Gustafsson, A., Johnson, M. D., & Roos, I. (2005). The effects of customer satisfaction, relationship commitment dimensions, and triggers on customer retention. Journal of Marketing, 69(4), 210–218.
Jones, T. O., & Sasser, W. E. (1995). Why satisfied customers defect. Harvard Business Review, 73(6), 88–99.
Lemon, K. N., & Verhoef, P. C. (2016). Understanding customer experience throughout the customer journey. Journal of Marketing, 80(6), 69–96.
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709–734.
Minto, B. (2009). The Pyramid Principle (3rd ed.). FT Prentice Hall.
Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship marketing. Journal of Marketing, 58(3), 20–38.
Oliver, R. L. (1999). Whence consumer loyalty? Journal of Marketing, 63(Special Issue), 33–44.
Reichheld, F. F. (2003). The one number you need to grow. Harvard Business Review, 81(12), 46–54.
Reichheld, F. F., & Sasser, W. E. (1990). Zero defections: Quality comes to services. Harvard Business Review, 68(5), 105–111.
Regulation (EU) 2016/679 (GDPR), Art. 5(1)(c),(e), 6(1)(f), 21(2)–(3), 28, Recital 47.
§ 7 UWG (Gesetz gegen den unlauteren Wettbewerb).

Task 1 · Level 1 · 30 min

Task 1 · Diagnostic Note · Kessler Präzisionstechnik GmbH

One task. Three levels. Level 1 establishes what the file records.

The case

Case assumption

TechSolutions GmbH (Stuttgart region, about 40 staff) delivers cloud migration and security projects to (mid-sized companies) manufacturers; projects run €60–150k. Kessler Präzisionstechnik GmbH (about 320 staff): TechSolutions delivered a migration project (contract value €118,000). was at Month 0, closed at Month +1, and no order followed. At Month +11 Kessler issued a tender (, formal tender) for a three-year service contract.

How to use this task

Each answer block starts with a FIND IT line: the route, the widget as printed on screen and the click that gets you there. Analyse in the app, then write in the answer area below it.
OBJECTIVE the file settles it. JUDGED your reasoning, in your words.
Five bins: Price, Trust, Benefit, Relationship, and Interpretation — not evidence. The sorting key is in Materi A · A6.
If two participants disagree, one has misread the file.
Time: about 10 min for the sort (1.1), 12 min for the map (1.2) and 8 min for the verdict (1.3).

Read first

Kessler journey map

Months run relative to go-live. The eight record markers sit at their months; a bar marks a record that covers a span. Select a stage to read its caption, or a marker to jump to its record.

What this showsSelect a stage (tap, click or press Enter) to read its caption, or a record chip below the axis for when it happened.

Block 1.1 · Sort the file

OBJECTIVE

FIND IT· Route 1 → Task 1 → “Kessler journey map” → press “Open all records”. Answer in the “Evidence board” below.

Analyse in the app. Write your result in the answer area below.

Draws on

Drag a record into a bin, or select it and then select a bin. Select a placed record to move it again.

Ctrl/⌘+Z · Ctrl/⌘+Shift+Z

Records not placed (8)

Select a record to read it here, then select a bin to file it.

Empty

Checks requested: 0Counted, not punished. It only outlines; it never gives the bin.

Evidence board

Live · redraws as you place records

Solid — an observed record
Hatched — an interpretation filed as if it were evidence
Outline — an observed record set aside

Block 1.2 · Map the customer

OBJECTIVEJUDGED

FIND IT· Route 1 → Task 1 → “Kessler journey map” and “Evidence board”, then “Loyalty map” inside this block. Answer in the grid below.

Analyse in the app. Write your result in the answer area below.

Draws on

For each row choose the bucket that says what the file shows, and the record it rests on. Where no record measures the thing, the honest bucket is “Not recorded”. Buckets: ● Positive · ◐ Weak · ○ Absent · ? Not recorded.

Construct (A2)

M1Satisfaction
A judgement of the outcome against expectation. Read it from a survey or a rating.
What the file shows
Record it rests on
M2Relative attitude (the attitude side of loyalty)
How the customer rates you against the alternatives. Typical measures: , stated intention, .
What the file shows
Record it rests on
M3Retention (repeat behaviour)
Observable continuation: renewals and repeat orders. Read it from what the customer did, not from what it said.
What the file shows
Record it rests on

Rope (A3)

M4Emotional rope
Trust and affinity: a named contact, contact cadence, sponsor continuity.
What the file shows
Record it rests on
M5Economic rope
Leaving costs more than staying. Needs both halves of the comparison: a rival's price gap and the customer's .
What the file shows
Record it rests on
M6Contractual rope
A formal obligation: a service or with a term and a .
What the file shows
Record it rests on

Positive: The file shows it in place, and good.
Weak: The file shows it, but thin or reduced.
Absent: The file shows it is not there.
Not recorded: The file has no record that measures it.

Checks requested: 0It outlines rows in amber and offers a question; it never gives the bucket.

Loyalty map · Kessler

What this showsPlaced from your readings M2 and M3. Nothing is placed yet: both axes are open.

Reading of the map JUDGED0 characters

One or two sentences. Frame: “The file places Kessler on the ___ axis as ___. It does not place ___ because ___. The record that would settle it is ___.” At least 30 characters.

Block 1.3 · Name the weak point

OBJECTIVEJUDGED

FIND IT· Route 1 → Task 1 → “Evidence board” and “Loyalty map”. Answer below.

Analyse in the app. Write your result in the answer area below.

Draws on

V1 · Category

The category holding the most observed evidence of a shortfall on TechSolutions' side.

PriceTrustBenefitRelationship

V2 · First cited record

Cite an observed record that supports your category. Not an interpretation.

V3 · Second cited record

A different record from V2, also an observation that supports the same category.

V4 · One sentence JUDGED0 characters

Sentence frame: “The file records ___ and ___. It does not record ___.” At least 20 characters.

Filing needs V1–V4 filled in — not correct. It unlocks a control in Task 2.

Your diagnostic note is still missing:

Preview of your note

File name: 1-participant-day1-l1-diagnostic.html