The problem keeps coming back: the iceberg, Lammsbräu and Shonaquip

For six months running, the operations lead at a small online homeware shop has hired six temporary packers for the last five days of the month. In month seven, late orders still pile up on exactly those days. The shop and every figure here are hypothetical. The iceberg model is meant for problems of this kind, and the post then reads two organisations with it: Neumarkter Lammsbräu in Germany and Shonaquip Social Enterprise (ShonaquipSE) in Cape Town.

Why does a problem keep coming back after you have fixed it?

Because the fix usually aims at the visible layer while the cause sits lower down. Goodman (1997) describes systems thinking as a move from observing events to spotting patterns of behaviour over time, then to the structures behind both. Changing structures that no longer serve us, mental models and perceptions included, widens the choices for a lasting solution (Goodman, 1997). He also gives a warning sign: when someone concludes that the problem is a need for more sales staff or more revenue, steer the conversation back to causes (Goodman, 1997).

How do the four layers apply to the hypothetical shop?

The hypothetical figures: 3,000 orders a month, of which 900 (30 per cent) land in the last five days, or 180 a day. The warehouse packs 150 a day, so five days clear 750 and leave 150 waiting, which is 17 per cent of 900.

Suppose the new bonus rule cuts the last-five-days share from 30 to 20 per cent. That is 600 orders, or 120 a day, below the 150 the warehouse can pack. It is a hypothesis to test with numbers: if the share has not fallen after two months, go back to the table.

To run it yourself, ask people from several departments to describe the problem together, since their mental models may differ a lot from yours, and open with what the team does not yet understand rather than who got it wrong (Goodman, 1997). Then write one line per layer and fix at the deepest layer you have the authority to change (this last step is my own).

Why would a brewery pay above-average prices and sign long contracts with about 170 farms?

The public pages do not explain the motive behind each decision. My reading is that Neumarkter Lammsbräu treats a future for organic farms as a condition of its own business. The facts come from two German-language pages of the company, read on 3 November 2026. See also the Lammsbräu case.

Where does the root change the solution? If organic raw material is treated as the purchasing department's job, the event-level answer is to find another supplier whenever stock runs short. The pages describe a different answer: a supplier community, with commitments on price and term. The 1977 goal came 12 years before EZÖB, so I guess that the structure followed the goal. That is inference. The pages do not say why EZÖB was founded, and they give no specific prices or farmers' views.

Why would a maker of children's wheelchairs train staff and lobby on policy?

Trafford et al. (2021) give the reason: devices can improve participation only to a limited extent if the wider environment is restrictive and stigmatising (Trafford et al., 2021). The paper is a reflection on practice. Three of its five authors are linked to ShonaquipSE, and they say it is not a formal evaluation. See also the Shonaquip case.

Where does the root change the solution? If the problem is children lacking seats, make and hand out more seats. If it lies in how the system supplies devices and how society sees disability, the answer also includes selling to public hospitals, training, parent networks and input on policy.

The accounts show financial strain. In the 2024–25 report, Shonaquip (Pty) Ltd's revenue rose 10.2 per cent to R43.1 million from R39.1 million, but its reserves of R1.518 million were only 37 per cent of the R4.1 million target. The Uhambo Foundation ran a deficit of R868 thousand, which the report attributes to direct cost allocation after cross-subsidy between the entities ended (ShonaquipSE, 2025). The report does not link this to the structure above, and neither do I.

What do the three questions of social innovation add to the iceberg?

The framework Three elements of social innovation asks whether a change deserves to be called social innovation. The library's framework page lists three elements: satisfying human needs that are currently unmet, reconfiguring social relations, and empowerment or political mobilisation of the people trying to meet their needs.

The first element fits ShonaquipSE better: postural support for children, against a background in which WHO and UNICEF (2022) estimate that more than 2.5 billion people need at least one assistive product (World Health Organization and United Nations Children's Fund, 2022). For Lammsbräu the pages describe a commercial product, so the framework fits less well.

When does the iceberg suit, when does it not, and what are the limits?

It suits problems that are chronic, familiar and already fixed several times. It does not suit a one-off incident, or a hunt for someone to blame (Goodman, 1997). Three limits:

• The model is a tool for asking questions, not proof. The two source articles are practitioner pieces and report no comparative results.
• The two organisations were chosen because they succeeded, and most of what is written about them comes from themselves or their insiders. I do not know of any organisation that built a similar structure and failed (survivorship bias).
• Reading structures from outside is inference. Neither organisation describes itself in these layers.

Going deeper: where does the model come from, and what did I read?

Goodman (1997) and Cunliff (2004) are two short pieces in The Systems Thinker. Goodman uses the iceberg as a prompt to describe a problem from three angles; Cunliff draws four levels and notes that authors draw from three to six (Goodman, 1997; Cunliff, 2004).

I read in full: both pieces, the two German-language Lammsbräu pages, the Trafford et al. paper and the ShonaquipSE report for FY2024–25. From the WHO and UNICEF report I read only the abstract, key messages and the need estimates. The three elements of social innovation come from the library's framework page, which names a 2013 chapter by Moulaert, MacCallum and Hillier as its source; I could not read it, so I do not cite it.