Optional foundation: vocabulary, numbers and evidence

A useful decision before a marketing plan

Marketing starts with deciding whom an organisation can help, what makes that help worthwhile and how people can find, choose and use it. Advertising is one possible activity. A charity that improves its booking process is doing marketing work even if it buys no advertising. A social enterprise that makes its buying terms clearer may help a commissioner choose responsibly without making a single social post.

This course uses one organisation throughout. You may use an organisation only with permission and without exposing personal or confidential information. Otherwise use the fictional community advice case. It is equally valid for every assignment and the capstone. No actual clients, donations, experiments or external messages are required. The audience is adults; this course does not establish arrangements for teaching children.

Know the roles

A service user receives a service; a beneficiary is someone intended to benefit, who may not be the immediate user. A payer finances it. A donor gives without buying the same service for themselves. A commissioner arranges provision for others. A volunteer contributes time. These roles can overlap, but their interests need not. A donor may like a moving story that the person featured finds intrusive. Good marketing cannot settle that conflict by counting donations alone.

An objective describes a change sought in a defined period. A tactic is an action taken to pursue it. A measure is an observation used to judge progress. An outcome concerns a change in people's situation or the mission, while an output describes something delivered. Sending 100 emails is an output. Ten people attending is a response or service-use result. Whether advice helps them resolve a problem requires further evidence. A baseline records the situation before a proposed change; it is not a promise about the future.

Work with the fictional advice service

The supplied fictional cohort contains 600 unique service-page visitors, 72 enquiries, 24 bookings and 15 attended appointments in one month. Monthly capacity is 40 attended appointments. A 90-day improvement project has at most £1,000 and 20 staff hours. These are invented teaching inputs, not a real charity's results or sector benchmarks. Unique visitors in this case are already defined as distinct within the cohort. Real tracking may not identify people accurately across devices or offline contacts.

A rate divides a relevant count by its denominator. Enquiry rate is 72 divided by 600, or 12%. Booking among enquirers is 24 divided by 72, or about 33.3%. Attendance among bookings is 15 divided by 24, or 62.5%. Attendance among visitors is 15 divided by 600, or 2.5%. These are different questions. Always write both the numerator and denominator before interpreting a percentage.

If attendance among bookings rose from 62.5% to 75%, that would be 12.5 percentage points. The relative increase would be 12.5 divided by 62.5, or 20%. Neither statement says that a change caused the increase. With 24 bookings, 75% would mean 18 attendances, three more than 15. Small count changes can look dramatic as percentages. Preserve counts and avoid implying a precision the observations do not support.

Capacity is not the same as demand. Forty appointment places minus 15 observed attendances gives 25 unused places in this month. It does not establish that 25 more eligible people want those exact appointments, that staff could handle a rush of enquiries or that every attendance would be useful. You need an account of the process before proposing more traffic.

Read a claim carefully

An observation reports something recorded. An inference interprets it. A hypothesis offers an explanation to test. A decision commits to an action under uncertainty. 'Fifteen people attended' is an observation in this fictional dataset. 'People cannot travel' is a hypothesis unless the case supplies evidence. 'Offer a different time' is a possible decision, not a finding. Keep these labels in your notes.

A useful source log has a claim, author, title, location, date, evidence type, limitations and the decision it informs. A survey may reveal reported reasons but omit people who never saw it. A randomised experiment may support a causal claim in one setting but not your setting. A textbook helps explain concepts; it does not establish this month's local barriers. Read methods and limitations before repeating an impressive headline.

When comparing arguments, write the strongest version of each. One colleague proposes promotion because the service has unused places. Another proposes improving booking because only a third of enquiries become bookings. Both could be partly right. The first needs evidence that eligible demand exists and can use the service. The second needs evidence that the loss reflects a remediable barrier rather than unsuitable enquiries or chosen alternatives. Better reasoning makes disagreement testable.

Learn safely without AI

You can complete the whole course with these text resources and a notebook. AI is optional. Generated text predicts plausible language; fluency is not a source check. A fictional output that says 'double social posts to increase attendance by 50%' has not supplied a study, mechanism or local test. Mark its forecast unsupported. Do not turn a generated persona or interview into evidence about real people.

If you choose a live tool, use only the fictional case or material specifically cleared for that use. Keep the prompt and output, identify the tool and date, check claims yourself and record substantive changes. Never paste client stories, contact details, sensitive information or internal documents just because a tool is free. If permissions or suitability are uncertain, use the manual route.

Build a small learning record

Create a local folder or paper workbook with a source log, assumption log, portfolio drafts and revision notes. Use descriptive filenames or page titles. Record what a number means, not just the number. Write dates on decisions. Remove names and identifying details from any authorised organisational material; removing a name alone may not make a distinctive story safe to share.

The foundation allocates 10 additional hours: 1.5 for this lesson, 1.5 for critical reading, 2 for numeracy and evidence lab, 3 for your starter brief, 1 for critique and 1 for reflection. The required course remains 160 hours without it. These are design allocations, not measured durations. Work in smaller sessions, use large text or read aloud if useful. No recording, visual worksheet, subscription or customer signup is required. This first authored version, dated 1 November 2026, has not received independent expert qualification or accessibility testing.

Lab — 2 hours

Calculate all four case rates without rounding until the final step. Explain the denominator for each. Calculate the number attending if 30 bookings had the same 62.5% attendance rate; explain why 18.75 is an expectation, not a possible observed count. Classify these statements: 15 attended; people dislike forms; the service should send reminders; capacity is 40. Check against the example after doing your work.

Starter brief — 3 hours

Write 500–700 words naming your organisation or the fictional case, the people it serves, a decision, a baseline and three unknowns. Add a source log with the case and one linked reading if accessible. If the reading is unavailable, compare the two internal positions in the resource and label the external source unread.

Critique — 1 hour

Circle every causal claim. Ask what observation supports it and whether another explanation fits. Check every percentage denominator.

Reflection — 1 hour

Rewrite the weakest claim and explain why the revised wording is more honest. Manual route: all activities use a notebook and the supplied output; live AI is unnecessary.

Self-review weights: evidence and source appraisal 25%; critical reasoning 25%; contextual fit 20%; feasibility and evaluation 20%; clarity and reflective judgement 10%. Score each criterion from 0 to 4 and calculate sum(weight × score / 4), out of 100. The optional foundation uses the common criteria but adds no weight to the assessed portfolio.

Evidence: use correct denominators, arithmetic and sources. Reasoning: separate evidence from assumption. Context: distinguish stakeholder roles. Feasibility: pose a decision the organisation could act on. Clarity/reflection: explain your revision and what remains unknown. A score of 0 is absent, 2 is partly supported and 4 is correct with limitations explained. Do not treat the total as an academic grade. Correct invented evidence or exposed personal information before proceeding.

Weak: 'Our 2.5% conversion proves our website is bad.' The denominator is identifiable but the cause is invented. Competent: 'Fifteen of 600 visitors attended; we should investigate the booking journey.' This connects an observation to an investigation without promising a result. Strong: 'Only 24 of 72 enquiries became bookings. We will first inspect response handling and eligibility information, because either could explain that loss; we cannot yet rule out low suitability or people's informed decisions not to book.' This names alternatives and a bounded next step. Numeracy check: 30 × 0.625 = 18.75 expected attendances, not a guarantee. For planning, show a range and capacity rather than claiming a fractional person attended.