Risk Disclosure

Investo publishes a research view on every equity listed on the Nigerian Exchange. This document sets out what those views are, how they are produced, and the specific ways they can be wrong. It is written to be useful rather than defensive — you cannot judge research you do not understand the limits of.

Nothing on Investo is investment advice or a recommendation to buy, sell or hold any security. We do not know your circumstances, your obligations, your tax position or your tolerance for loss, and our outputs are not tailored to them. Investo does not hold your money or execute trades — see our Terms of Service. Where a decision matters to you, consult a professional licensed to advise you.

The value of shares and the income from them can fall as well as rise. Past performance does not predict future returns. A company can lose its entire equity value, and shareholders rank last if it is wound up. Nigerian equities have historically been volatile, and a naira return is not a dollar return: currency movement alone can turn a naira gain into a loss in another currency.

Never invest money you cannot afford to lose, and never invest borrowed money on the strength of a model output.

A verdict is the output of a deterministic valuation engine and a language model reading the company's own filings. It is an estimate under stated assumptions, not a finding of fact and not a forecast. It carries:

• A direction — buy, accumulate, hold, watch or avoid — reflecting how the price compares to our modelled value on that day.
• A confidence band — high, medium, low or insufficient — scored on how much clean annual reporting stands behind the name, how well the valuation method fits it, how recent the evidence is, and whether the company has characteristics (holding-company structure, FX distortion, recapitalisation) that make the standard methods less reliable.
• A fair-value range, sometimes. Below the medium band we publish a direction and our reasoning but no fair-value range, because we do not think the evidence supports a number. Where a computed value has to be clamped to a credibility limit, or where our valuation methods disagree with each other by more than we can defend, we withhold the point estimate entirely.

Confidence is a statement about our evidence, not about how likely the share is to rise. A high-confidence "avoid" means we are confident in the analysis, not that the price will fall.

Figures are extracted from filings published by the exchange, using automated reading including AI vision on scanned documents, and are supplemented by third-party market-data vendors. This works well and it does fail. Failures we have found in our own data, and now guard against, include:

• Unit and scale errors. Nigerian statements are printed in thousands or millions, with the unit in a column header. When that header is misread the figure is plausible and wrong by a factor of a thousand. In August 2026 an audit of our corpus found 44 such figures across 13 companies. They have been removed, and every new figure is now checked against that company's own history before it is stored, with the whole corpus re-checked daily.
• Mis-mapped lines. "Revenue" can be bound to a sub-total rather than the top line, which distorts every margin derived from it. Where we detect a figure that cannot be right — profit above revenue, equity above assets — we now delete the figure rather than publish it.
• Vendor ratios. Third-party ratios are computed arithmetically without asking whether the denominator is meaningful. A company whose equity has been destroyed by currency revaluation can show a return on equity of several hundred per cent, which is true division and meaningless analysis.
• Gaps. A company that changes its financial year, or files late, may have no comparable annual period at all. We flag this rather than fill it in.

We correct errors when we find them, and a corrected figure changes the verdict that stood on it. Verdicts are revised without notice.

Prices shown are exchange closing prices, not live quotes, and a verdict is computed against the price at the moment it was generated. That price is already historical when you read it. Verdicts are refreshed on a schedule, so the one you are reading may predate a result, an announcement or a price move that changes the case entirely.

Every verdict carries the date it was generated and the price it was generated against. Check both before acting on it.

Some risks are properties of the market rather than of any analysis, and no research product can remove them:

• Liquidity. Many listed Nigerian equities trade rarely and in small size. A quoted price is the price of the last trade, not a price at which you can buy or sell any quantity you like. Exiting a position in a thin name can take days and can move the price against you. This is frequently the dominant risk in a small cap, and it is unrelated to whether the company is good.
• Free float and concentration. Where most of a company is held by a parent or a small group, the traded portion is small and the price is easily moved.
• Currency. Naira devaluation affects reported earnings, balance sheets and the real value of returns, and has repeatedly done so at scale. Companies with foreign-currency obligations can swing between profit and loss on translation effects alone, with no change in the underlying business.
• Disclosure timeliness. Filings can be late, restated or incomplete. Our analysis can only be as current as what the company has actually published.
• Macro and policy. Interest rates, inflation, fuel and electricity costs, capital controls and regulatory change can affect every holding at once, in ways no company-level analysis anticipates.

Our written analysis is produced by a language model working from data we supply. The model does not originate the figures — the numbers come from filings and computation, and the model is constrained to the evidence in front of it. It can still misread context, weigh a factor poorly, or express more certainty in prose than the underlying data supports. Ask, memos and thesis text are subject to the same limits.

We publish our confidence, our valuation method, the number of reporting periods behind each view, and where our own committee disagrees with the model, precisely so you can discount our output where it deserves discounting. Use those signals. A verdict without a fair-value range is telling you something.

We make no representation that Investo is accurate, complete or fit for any particular purpose, and we do not promise any investment outcome. Coverage, data and features change. Nothing here creates an advisory or fiduciary relationship between you and One Digital NG, and we are not responsible for decisions you make using our research. Our liability is limited as set out in the Terms of Service.

Investo never asks you to send investment money to us, and has no mechanism for accepting it. We do not manage funds, guarantee returns, run investment schemes, or operate groups promising profits. Anyone using Investo's name to solicit investment funds is committing fraud. Report it to femi@onedigital.ng and to the Securities and Exchange Commission.

If a figure on Investo does not match the filing it claims to come from, we want to know — it is the fastest way we find the errors described in section 3. Write to femi@onedigital.ng or use the contact form, and tell us the ticker, the period and the figure. We publish a live coverage report so our claims about what we hold can be checked against the data itself.