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-day9-l1l2-real-time-file.
Route 1 · Levels 1 and 2 · Knowledge and application
Real-time retention: answer in the moment, personalise with sense, and prove it works
Materi A · 60 minutes, facilitator-led
Customer retention in real time: why speed matters, where to respond and personalise, and how to measure it
Seven cards, Level 1 and Level 2 in one run: knowledge first (what a delay costs, personalisation in the moment, where to respond at once, what speed is worth), then application (KPIs in real time, testing and feedback loops, choosing measures). Every diagram uses Neckar Hosting, another provider, so the task is never answered for you.
OPTIONALA1 · Why real time matters: what a delay costs8 min
Deepens a card a Core task block already covers. Not needed to complete the Real-Time Analysis File.
A2
Personalisation in the moment: value or intrusion
CORE9 min
in the moment changes what a visitor sees now, from who they are or what they do: a , an offer, the first answer on the page. It is felt as value when it uses what the visitor gave you, and as intrusion when it shows you watched them.
In plain words
Personalising the moment means the page fits the person looking at it: the customer who logged in, the visitor who came from a campaign. It helps when it uses what they told you. It creeps people out when it shows you were watching.
Three visitors, three levels of personalisation · a worked example on Neckar Hosting
The pointA page can change what a visitor sees from what the visitor gave: the ad they clicked, the account they logged into. That is felt as help. Using more than they gave, or asking before showing anything, is felt as intrusion.
3 short steps in plain words. The picture changes as you press Next. Or skip this and use the buttons below on your own.
First visit, from a search
The standard page, with the three questions first-time visitors ask most.
From the law-firm campaign
Hosting for law firms first: client confidentiality, a case from a Mannheim firm.
Logged-in customer
Their servers' status, their next renewal date, and the add-on that customers of their size added next.
What this showsIn plain words: From the law-firm campaign · Fitting: The page uses what the visitor gave: the campaign they clicked, the account they logged into. It helps them decide faster, so it is felt as added value.
A worked sort: three ideas at Neckar Hosting
“Answer every chat on the pricing page within two minutes.”
“Show law-firm visitors the confidentiality page first.”
“Test two versions of the order form for three weeks and keep the better one.”
Illustration on Neckar Hosting (Case assumption). A dashed frame marks personalisation that goes too far.
Choose a visitor and a level and read what the page shows and how it is felt; then try the worked sort below.
OPTIONALA3 · Where to respond at once: chatbots, callbacks, social media and adaptive content9 min
Deepens a card a Core task block already covers. Not needed to complete the Real-Time Analysis File.
OPTIONALA4 · What speed is worth: closing rate, lift and extra revenue9 min
Deepens a card a Core task block already covers. Not needed to complete the Real-Time Analysis File.
OPTIONALA5 · KPIs in real time: outcome, driver, guardrail and vanity metrics8 min
Deepens a card a Core task block already covers. Not needed to complete the Real-Time Analysis File.
OPTIONALA6 · A/B testing in digital channels and feedback loops9 min
Deepens a card a Core task block already covers. Not needed to complete the Real-Time Analysis File.
A7
Prioritising measures: effect, speed, scalability
CORE8 min
Choose measures by the plan's three tests, each Low (1) to High (3), multiplied: effect (how much it moves the result), speed (how soon it works) and scalability (does it reach every visitor without extra cost). Then check the budget and which problems you answer.
In plain words
When time is short, how soon a measure works matters as much as how well. Score each option: how much it moves the result, how quickly it works, and whether it reaches every visitor without extra cost. Multiply the three.
Three measures of Neckar Hosting, scored
The pointScore a measure on three questions: how much does it move the result, how fast does it work, and does it reach every visitor at no extra cost? The three scores are multiplied, so one weak answer lowers the whole.
3 short steps in plain words. The picture changes as you press Next. Or skip this and use the buttons below on your own.
What this showsIn plain words: on the order page (€20,000, 3 weeks): effect 3 × speed 3 × scalability 3 = 27. It is working within 4 weeks, so speed is 3. It answers visitors where they decide, works within a month and serves every visitor once built.
Choose a measure to read its three scores and why each one is what it is.
References · Materi A (15)
Check every source before you teach from it: page numbers and editions differ between printings.
Oldroyd, J. B., McElheran, K., & Elkington, D. (2011). The short life of online sales leads. Harvard Business Review, 89(3), 28. (Leads contacted within an hour are far more likely to be qualified than leads contacted later.)
Aguirre, E., Mahr, D., Grewal, D., de Ruyter, K., & Wetzels, M. (2015). Unraveling the personalization paradox: The effect of information collection and trust-building strategies on online advertisement effectiveness. Journal of Retailing, 91(1), 34–49. (Personalisation helps when customers see how their data is used; covert collection makes it feel intrusive.)
Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of social media. Business Horizons, 53(1), 59–68. (Social media as a two-way channel in which firms must listen and respond.)
Markey, R., Reichheld, F., & Dullweber, A. (2009). Closing the customer feedback loop. Harvard Business Review, 87(12), 43–47. (Act on customer feedback quickly and report back; the loop is what improves the service.)
Davenport, T. H., & Ronanki, R. (2018). Artificial intelligence for the real world. Harvard Business Review, 96(1), 108–116. (Start from a business problem with small, measurable projects, not from the most ambitious technology.)
Peppers, D., & Rogers, M. (1993). The One to One Future: Building Relationships One Customer at a Time. Currency Doubleday. (Treating different customers differently, from what each one does.)
Adam, M., Wessel, M., & Benlian, A. (2021). AI-based chatbots in customer service and their effects on user compliance. Electronic Markets, 31, 427–445. (What chatbots do well in first contact, and how their design affects customers.)
Huang, M.-H., & Rust, R. T. (2021). Engaged to a robot? The role of AI in service. Journal of Service Research, 24(1), 30–41. (Machines take routine tasks; feeling and judgement stay with people.)
den Boer, A. V. (2015). Dynamic pricing and learning: Historical origins, current research, and new directions. Surveys in Operations Research and Management Science, 20(1), 1–18. (How prices are set by rules that learn from demand.)
Provost, F., & Fawcett, T. (2013). Data Science for Business. O'Reilly. (Rates, lift and expected value as the basic tools for reading a result.)
Kaplan, R. S., & Norton, D. P. (1992). The balanced scorecard: Measures that drive performance. Harvard Business Review, 70(1), 71–79. (A few linked measures, results and the drivers behind them, instead of many unrelated ones.)
Ries, E. (2011). The Lean Startup. Crown Business. (Vanity metrics against actionable metrics; learning through controlled experiments.)
Kohavi, R., Tang, D., & Xu, Y. (2020). Trustworthy Online Controlled Experiments. Cambridge University Press. (Random split, one change, a size fixed in advance, guardrail metrics, and the danger of stopping early.)
Hubbard, D. W. (2014). How to Measure Anything, 3rd ed. Wiley. (Start from the decision; measure what would change it.)
Regulation (EU) 2016/679 of the European Parliament and of the Council (General Data Protection Regulation), Art. 6 (lawful basis), Art. 21 (objection to direct marketing) and Art. 22 (automated individual decisions).
Task 1 · two core blocks, one per level; optional blocks folded
LiveConnect IT Services GmbH sells managed IT and cloud services to the . Its website draws many visitors, but most leave within seconds, few start a conversation, and a quote request waits a day or more for an answer. Chat, e-mail, campaigns and are run by different teams, and nobody coordinates them or knows which of them works.
One first sign: last quarter 900 quote requests were answered after more than a day and closed at 8%, while 300 answered within one hour closed at 24%. Fast answers closed 3 times as often, but sales chose which requests to answer fast, so it is a hint, not proof.
What you have
Nine real-time ideas from LiveConnect's teams (Block 1.1).
Eight moments on the website (Block 1.3); last quarter's speed figures (optional Block 1.2).
Twelve metrics LiveConnect reports today (Block 2.1) and six measures it could fund (Block 2.4).
The limits
Budget: €170,000
Time: 4 months
The cost and weeks of every measure are printed in Block 2.4.
How the task runs · two core blocks, about 20 min
Block 1.1: sort nine real-time ideas into respond, personalise or learn, and name an opportunity of your own (Level 1, material A2).
Block 2.4: choose three measures, score them and defend the order (Level 2, material A7).
Six more blocks (about 44 min) are optional and folded.
Case assumption
The brief says: high visitor numbers, low and few closings; high , low interaction and measures not coordinated; €170,000 and four months. Everything else is made up for this exercise: the pages, the speed figures, the metrics, the rates and the costs.
Part 1
Understand the real-time effect
Level 1 · Knowledge
Block 1.1 · Respond, personalise or learn?
OBJECTIVECOREabout 6 min
FIND IT· Route 1 → Task 1 → the nine real-time ideas on the sort board below, from LiveConnect's service, sales, marketing and product teams. Answer on the sort board.
Analyse in the app. Write your result in the answer area below.
Draws on
Drag an idea onto a lever, or select it and then select a lever. Select a placed one to move it again. One lever per idea: the one the idea mainly pulls.
Ctrl/⌘+Z · Ctrl/⌘+Shift+Z
ideas not sorted (9)
Empty
Empty
Empty
Checks requested: 0
A check counts how many placed ideas hold. It never says which, because naming the wrong ones would name the answer.
0 / 30 characters
Name a moment in LiveConnect's customer contact, what happens there today, and what a reaction would change (“so …”). At least 30 characters.
Still missing in Block 1.1
OPTIONALBlock 1.2 · Read the speed figures: two closing rates side by side5 min
Practises reading one comparison without being fooled by it (sales chose which requests to answer fast); the choices of Block 2.4 do not need it.
OPTIONALBlock 1.3 · Where to respond at once, where to personalise, and three improvements9 min
Develops three simple approaches from the customers and moments of the case; Block 1.1 and the measures of Block 2.4 are answered without it.
OPTIONALBlock 1.4 · Coaching reflection: from Level 1 to Level 25 min
A reflective bridge between Level 1 and Level 2, not content the Real-Time Analysis File itself needs.
Part 2
Make it measurable and choose
Level 2 · Application
OPTIONALBlock 2.1 · Tag LiveConnect's twelve metrics by kind, and name your three KPIs11 min
Practises telling an outcome, a driver, a guardrail and a vanity metric apart; the choices of Block 2.4 are made and scored without it.
OPTIONALBlock 2.2 · What each kind of metric is worth, and the uncertainties6 min
Reads what each kind of metric tells management, from your tags in Block 2.1, and what can mislead a measurement; Block 2.4 can be answered without it.
OPTIONALBlock 2.3 · Design a fair A/B test8 min
Applies the fair-test rules of Materi A6 to a chat on the pricing page; the measures of Block 2.4 are chosen and scored without it.
Block 2.4 · Choose three measures, score them, put them in order
OBJECTIVEJUDGEDCOREabout 14 min
FIND IT· Route 1 → Task 1 → “The limits” in the case above (€170,000, 4 months) and the six measures below. Answer by choosing three and filling their cards.
Analyse in the app. Write your result in the answer area below.
Draws on
Choose exactly three of the six measures. Together they have to fit €170,000 and 4 months (16 weeks). Prices are Case assumptions: each is built from the parts printed on its card.
Which problem a measure answers is for you to see: tick it and watch the picture under the cards. The weeks on a card decide its speed score (rule in the box below). and scalability are your judgement.
0 of 3 chosen.
What your plan does to the three problems
○ High bounce ratesVisitors leave after a few secondsNot answered yet
○ Low interactionVisitors do not ask or clickNot answered yet
○ Measures not coordinatedNobody sees what worksNot answered yet
When each measure starts working (16 weeks = 4 months)
0481216
No measure chosen yet.
Hatched = being built, nothing changes for the customer yet. Solid = working.
What this showsIn plain words: Choose a measure and the three boxes show which problem it answers; the bars show when it starts working.