Engagement & Standards

Periplus is built for the question that remains after the record has been read: where the decisive fact was never written down, and being wrong is costly enough to warrant reaching it.

The questions take a recognisable form.

• When does a signature bind, and when is it ceremonial?
• Who really holds a decision that formally sits elsewhere?
• When is consent real, and when has it only been recorded?
• When is a licence secure, and when can it be revoked once the official who granted it has moved on?
• Who stands behind a counterparty, and will they perform once relationships replace documents?
• When is a clean structure politically contaminated?
• When is a court victory collectable, rather than blocked by whoever protects the asset?

The decisive fact in each is spoken only in private, and in its own language—held in-country, at the centre as often as in a province, and seldom in full by any single person—and reached only in conversation. In the firm’s jurisdictions these are not rare problems: they attach to the ordinary decisions of these markets—an entry, a licence, a partner, an acquisition—because there the decisive fact is routinely unwritten. Some questions run forward, to a decision still forming, where knowing early is the difference between acting and reacting. Others run backward, to conduct already contested, where what happened must be established against an interested account.

Where a question can be settled from public sources, by a desk analyst, or by standing subscription coverage, Periplus says so; scoping is at the firm’s cost. The work is reserved for the problem that requires it.

Track II is the diplomatic name for dialogue held outside official channels: conversations between people who speak for themselves, to someone they trust, not for any government or institution. The tradition exists because official channels produce positions rather than intentions, and candour survives only in settings that are unofficial and held in trust; its norms—confidentiality, independence, non-attribution—exist to keep them so.

Commercial matters have the same structure. A company facing a government or a counterparty has its own official channels—legal advisers, bankers, the negotiating table—and its own formal record. The decisive facts sit outside both: who is actually deciding, whether consent will hold, what a counterparty intends. Periplus works in this tradition because that is where such facts live, and its role is the tradition’s role: to inform the official process, never to conduct it. The firm does not negotiate on any party's behalf or seek to influence official decisions. Where separately instructed, it may support a client's own overt, authorised engagement with a government, openly and in the client's name; that support is distinct from its collection work.

The firm’s own formation lies inside the tradition. The practice behind Periplus took shape reading the Ethiopian civil war for foreign ministries and investors during a communications blackout; the capability it built went on to serve Track II peace processes in Africa and the Middle East and, behind the scenes, the top-tier strategic intelligence and disputes advisory firms. That commercial work has since run in all but two of Africa’s 54 countries: hundreds of mandates spanning subsea cable transit rights, telecoms market entry, upstream M&A positioning, mining portfolio monitoring, and cross-border concession liability. Periplus was formed to offer it directly. The tradition’s norms bind that history as they bind every mandate: the processes, the firms, and the parties are not named.

In transactions, usually before terms are signed, when the question is whether the counterparty and the operating environment are what they appear to be. In contentious matters, instruction often comes after the legal team is in place but before pleadings are filed, while the underlying picture still has to be established. Standing engagements run across the life of an asset, position, or dispute, cadenced to political and commercial cycles; they work alongside the client’s own government-affairs, strategy, and risk functions, supplying the collection those teams cannot reach, for assessments that remain theirs. The range runs from pre-investment scoping to active enforcement; the common factor is a decisive fact still to be established.

For investment and transaction mandates: written analytical reports and oral briefings, framed around the decision at hand — a reading of a counterparty and who stands behind it, an assessment of whether a structure will hold, a basis on which to proceed, reprice, restructure, or walk.

For standing engagements: source reporting as it arises, with immediate notice when the picture moves, and periodic assessments that bring the reporting into a current reading — what has changed, what now holds, and what remains to be tested.

For matters led by the client's legal team: chronologies, corroboration notes, and intelligence assessments that inform case strategy and identify the evidence the legal team can develop — the witnesses, documents, and verifiable facts for proceedings, delivered to the procedural timetable. Source reporting informs strategy and identifies the evidence counsel can develop; whether any material can be used in proceedings depends on the applicable rules and is for counsel to determine.

For enforcement and recovery, and for funders underwriting it: a collectability assessment — whether value can actually be reached, who would obstruct it, and the realistic path to it — before the cost is authorised.

Where separately instructed and within demonstrable expertise, expert input scoped to the requirements of the proceedings. Every deliverable separates what is established from what is corroborated from what remains unresolved.

A standard single-jurisdiction assignment, typically six to eight lines of enquiry over ten working days including a round of follow-up with sources, runs three to five weeks from scoping to delivery. A narrower rapid assessment can be scoped against an immediate decision — a counterparty question ahead of signature, a sudden move against an asset — and delivered in days, with its coverage and limits agreed in advance. Wider mandates run over a period agreed at scoping. There is no cold start: the networks are standing, so where they cover the relevant jurisdiction and sector, work begins on instruction. Retainer arrangements keep the network on standing activation, with no scoping period at all.

Over years, not per mandate. The relationships were formed in the course of long immersion in these regions, with interlocutors who are of the places they report from—inside these worlds long before any mandate arrives—and many predate the firm itself. Each is vetted and brought into a structured framework with defined confidentiality, anti-corruption, and reporting obligations; the relationships are held directly by Periplus, not subcontracted through intermediary firms.

Composition matters as much as reach. Power, information, and trust run along ethnic, regional, linguistic, and communal lines as much as through formal institutions, and a network drawn from a single community or a capital elite cannot see across them. Periplus composes its networks to reflect the actual structure of a society—across its regions, its languages, and the affiliations through which things are really decided—so that what it reads is not the view from one vantage but from several.

Interlocutors are treated as long-term relationships, not a resource drawn on and discarded: engaged fairly, kept in contact between mandates, never exposed for a single piece of work. Many speak at some risk to themselves, and the firm’s first obligation is to them. Trust kept over years is what produces candour—and what lets the firm return to the same person as a situation develops, to test a reading against what is already known. Understanding of this kind compounds; a contact used once and dropped is lost.

The firm does not detail the structure itself. Protecting how the network is organised is part of protecting the people within it.

For each mandate, between two and six in-country interlocutors operate independently within their own professional networks, at the levels the question reaches — capital ministries, sectoral regulators, provincial administrations, and the actors around them. Each reports without sight of the others, and the firm weighs the fragments against one another, testing them for agreement and divergence and checking, before two accounts are treated as corroboration, that they do not rest on the same underlying source. A claim that holds across sources from different circles is worth more than one resting on a single account. Sub-national reach is inherent to the model: the provincial level is often where the decisive variable sits.

Networks are pre-established across 85 jurisdictions in Africa, the Middle East, Central Asia, South Asia, and Southeast Asia. They extend to the sub-national and provincial levels where regulatory decisions are executed, where community consent is won or withheld, and where political dynamics play out in practice—the provincial governor’s discretion, the regional regulator’s reading of a rule, a district administration’s posture toward a project. This is the level at which risk takes shape, and above which national-capital analysis usually stops. Coverage is not claimed where it does not exist. Mandates are accepted only where Periplus holds embedded relationships in the relevant jurisdiction and sector—confirmed at scoping, before any commitment is made.

Because the integrity of the reporting depends on it. Collection is carried out without sight of the client or the client's interest: interlocutors report on the question itself, not knowing who is asking or what answer would suit them. Reporting produced that way is far harder to bend, consciously or not, toward a desired conclusion, because no desired conclusion is in view. It is then tested: across interlocutors from different circles, for agreement and divergence; for whether two accounts rest on the same underlying source; against contrary evidence; and against each source's knowledge and interests.

The analysis brings the client's actual question to bear on what has come in. The principal, who holds the client's interest and did not collect the reporting, weighs it and forms the conclusion. The client's purpose shapes what is asked and how the findings are read, never what the sources reported in the first place. Commissioned intelligence tends to drift toward telling the client what the client wants to hear; this structure is built to resist that, and what the firm reports is what survives the separation, with the basis and limits of each material finding made explicit.

This is the practical translation of Track II independence into a commercial setting. It is reinforced by conflicts cleared at firm level before any mandate is accepted, refusal to act for rivalrous interests in the same matter, principal-level accountability for the final judgement, and the long-term holding of both client and network relationships rather than transactional volume.

No. The knowledge is oral, and it is dispersed—held in pieces by many, in whole by none. So the work is conversational: a question is defined with the client and broken into sub-questions; interlocutors carry each into their own networks, in person and in the relevant language; and the fragments that come back are assembled into a single reading.

Open-source research, however thorough, retrieves what has been recorded: filings, registries, databases, and the press. Periplus works where the decisive knowledge was never recorded—and, often, was never designed to be legible to outsiders. It is the knowledge a place keeps largely for itself: obligation and reputational debt, factional balance and informal protection, what a minister intends as against what the communiqué states, which relationships govern a decision no chart shows. It is disclosed only to someone trusted enough to be told.

Periplus draws political ethnography, regulatory and strategic analysis, and field intelligence into a single discipline, and the reason is that this knowledge is rarely reached by an expert advising remotely, an analyst reading documents, or a consultant in country for a week. Each reads a place from outside, through a framework brought to it — an etic account, in the anthropologist's terms. Periplus works from within — an emic understanding — through interlocutors already part of the world they report from, who read it in its own terms. And presence has to be local rather than national: a capital and its provinces are often separate worlds, and a contact in the capital can read a distant province little better than a foreigner can.

Public reporting records what has happened. Periplus reports what is forming, and what it means, from inside the institutions and networks where decisions are made.

AI can accelerate the reading of the record; it cannot create the record that was never made. Periplus uses it accordingly — for what has been digitised and can be reasoned over — and is built on what lies outside that boundary, in two directions at once. Some knowledge was never recorded: it is held by particular people and disclosed only within trust built over years, in person and in the relevant language — something a model can analyse but cannot be party to, because it cannot be the person taken into confidence or sit where the matter is decided. And some was recorded but never digitised: the paper file in a provincial registry, the archive no system indexes, the record kept deliberately off any network.

In frontier and emerging markets that boundary encloses much of what matters. Less has been digitised, the decisive arrangements are often kept off the record, and the deeper a question runs, the less of it a model has to work on, however capable it becomes. As more of the digitised record yields to machines, the value moves toward what was never digitised, or never written. The firm is built on the part of the work that grows more valuable as the machines improve.

The value of that material still depends on its provenance, its corroboration and its interpretation, and Periplus combines it with documentary, structured and geospatial sources where these are relevant. Source reporting and client material are not placed on external systems; analysis of the digital record is kept separate from the human-source reporting it is read against.

Periplus’s findings are intelligence assessments, not adjudicated facts. Each deliverable marks what is established, what is corroborated, what rests on a single source, and what remains open, with a confidence level throughout. They inform a decision or a case strategy; they do not replace the client’s own judgement or their lawyers’ advice. Where findings will carry real weight—relied on in proceedings, or stood behind in a transactional warranty—their use is best worked out with the legal team.

Findings are presented as they stand, including where they are inconclusive, contradict working assumptions, or reveal what was not anticipated. Where sources diverge, this is noted; where a finding rests on a single source, this is stated. The finding that cuts against what a client expected is often the most useful one it receives—and it is worth most while a position can still change.

Yes, depending on the material and the applicable rules. Some reporting informs strategy or identifies lines of enquiry; other material may be developed for use as documentary evidence, witness testimony, or expert analysis. An intelligence assessment is not automatically proof of the events it describes. Admissibility and evidential weight are matters for the relevant court or tribunal.

Where potential evidential use is contemplated, Periplus agrees the scope and collection protocol with counsel at the outset. This covers the factual questions, provenance, recording and preservation of accounts, source consent, and possible disclosure or questioning. Reporting distinguishes direct knowledge from information received from others and from analytical inference, and records corroboration, contrary indicators, and limitations.

Confidentiality, source protection, and privilege are addressed with counsel before collection. Whether privilege attaches depends on the applicable law and the circumstances of the engagement; lawyer involvement or a confidentiality agreement does not establish it by itself. Permission to speak confidentially is not treated as consent to be identified or to give evidence.

Interview records and supporting material are preserved under agreed retention and security protocols. Any expert report or testimony is separately scoped to the expert’s qualifications, independence, and the requirements of the proceedings.

Yes—alongside, not in place of. Periplus does not duplicate or displace advisory work already under way. In lawyer-led matters, findings are shared with the legal team rather than the client directly, and sequenced around litigation and transaction timetables. The firm does not require exclusivity.

With client consent, yes. For committee-level matters involving multiple institutional parties, Periplus can structure an engagement to serve a group of principals under a single confidentiality framework; and where the client determines that sharing serves the mandate, deliverables can be passed to co-investors or committee members under the same framework.

Client identities are held exclusively by Periplus principals and are not disclosed to interlocutors or any third party, except where the client has instructed overt engagement in its own name or has consented to a specific disclosure. Collection, coordination, and analysis are structurally separate. No engagement detail is disclosed to anyone without express client consent.

In intelligence reporting, source identities are not disclosed to clients. Findings are presented as assessments, not attributed statements; where a source's institutional role is material to the credibility of a finding, it is described by category — a senior official within the relevant ministry, a serving regulatory executive — without identification. Where a mandate contemplates evidential use, that is a separately agreed process: the source is consulted before any step is taken, their consent governs, and the arrangements for consent, disclosure and protection are agreed with counsel in advance. Permission to speak confidentially is never treated as consent to be identified or to give evidence.

No payments of any kind are made to sources for information; intelligence is given on the basis of trust built over years, not on transaction. A five-stage anti-bribery process applies to every engagement: pre-engagement training and contractual prohibitions; written attestations with each research submission; quality review against public sources; additional verification on high-risk engagements; and payment withheld until all stages are complete. Where a mandate requires documents—archival records, corporate filings, official publications—Periplus obtains them only through lawful channels: licensed lawyers for formal registry requests, accredited representatives of publishers for archival retrieval, direct purchase of published material at market rates. Every procurement carries a provenance statement recording its source, how it was accessed, what was paid, and to whom—available to counsel where chain of custody or evidentiary standing is in question.

Periplus does not conduct surveillance, intercept communications, access systems without authorisation, seek to obtain privileged communications, misrepresent the identity of its interlocutors, or make payments or other inducements to public officials. The firm does not engage in any activity requiring authorisation under the Regulation of Investigatory Powers Act 2000 or the Investigatory Powers Act 2016.

Each mandate is screened for sanctions exposure before acceptance, and interlocutors are screened against UK, US, EU, and UN lists before deployment. Where the matter touches comprehensive sanctions regimes, additional review applies—secondary-sanctions exposure, end-use restrictions, and any licensing requirements—before any in-country work begins. Periplus does not act for, or accept instruction from, designated persons or entities, and does not facilitate transactions in their favour. The Compliance Protocol is available on request.

Periplus does not engage current public officials. Every interlocutor must confirm they hold no current role as a public official and disclose any active government contract that could affect their independence; former officials are subject to enhanced due diligence covering conflicts, ongoing access, and post-employment restrictions. Interlocutors may speak with current officials, but only within their own pre-existing relationships: Periplus does not introduce them to officials, and does not fund the relationships through which such conversations happen. Those conversations are voluntary and carry no payment or inducement; what is said is reported as a third-party account, and never used to facilitate a transaction with the institution concerned.

Each mandate is screened for conflicts before acceptance. The firm does not act for opposing parties in the same matter, or where knowledge from a prior engagement would create an informational conflict between current parties. Conflict checks cover principals and the relevant segment of the interlocutor network. If a conflict surfaces mid-engagement, work is paused and disclosed to the affected client; it resumes only with informed consent, and where consent is not appropriate, the firm withdraws.

Periplus declines mandates it cannot answer through human-source dialogue, and matters where the decisive variable sits beyond any network’s reach. It declines where access does not exist in the relevant jurisdiction, where the timetable is too compressed for considered work, or where a conflict cannot be resolved. And it declines on principle where the client’s aim is incompatible with the firm’s independence or compliance obligations, or where the work would serve to confirm a predetermined conclusion. Where a question is real but Periplus is not the right firm for it, it refers the client elsewhere.

With a conversation, in confidence and without obligation; where preferred, a non-disclosure agreement is signed before anything is described—the firm’s or the client’s own. The prospective client sets out the situation; Periplus forms a preliminary view of whether the question is one it can answer, and says so if it is not. Scoping follows at the firm’s cost: the question is defined, coverage in the relevant jurisdiction and sector is confirmed, and conflicts are cleared. What comes back is a short proposal: the question as understood, the approach, the timeline, and the terms. The shape of delivery is agreed there too—interim briefings as findings develop or a single report at the end, delivered in writing, as an oral briefing, or both. On signature of the engagement letter, the brief is settled together—the core question broken into the sub-questions the network will carry—and collection begins; the raw material of every finding is source conversation.

Fees are agreed at the outset of each mandate—a fixed engagement fee, a retainer, or a combination—set by the scope and complexity of the work. Periplus does not bill by the hour, and no part of its compensation depends on findings reaching any particular conclusion.

Engagement records are retained under agreed retention and security arrangements, according to legal, regulatory, and privilege requirements. Routine reporting is held for a defined period and then deleted, unless needed for compliance, an ongoing matter, or chain of custody; for lawyer-led work, retention follows the legal team's direction. Clients may request earlier deletion, except where the firm is legally required to preserve a record. Where Periplus is later re-engaged on a related matter, the retained material remains available under the original confidentiality and privilege framework.