Start a project

Turning data, opinion and markets into intelligence for better decisions.

Independent research, polling and social intelligence for the companies, institutions and partners who need to know what is happening, and what to do about it.

Start with your objectiveSee the intelligence products
3,456 public comments, coded by hand. Somali, English, Arabic.
We readBusiness insightsTelecommunicationsConsumer behaviourDevelopmentPublic policyInvestmentAll sectors

Research tells you what people say. Intelligence tells you what it means.

Every engagement is built backwards from the client's objective. We say what the evidence can and cannot support, separate finding from interpretation, and end with what to do next. Information is the input. The decision is the product.

DataWhat was said, bought, answered or counted.
FindingWhat the evidence shows, with its limits stated.
InsightWhy it is happening, in the Somali context.
ImplicationWhat it means for your market, programme or policy.
RecommendationWhat to do next, in order, with the trade-offs.
DecisionA choice you can defend to a board, a donor or a minister.

Seven things we do, and the question each one answers.

We do not sell methods. We answer the question a senior team cannot answer from the inside.

Market intelligence

Is this market worth entering, and on what terms?

Market sizing, competitor profiling, pricing and distribution, barriers and white space, read through the Somali operating environment.

Consumer and market research

Who buys, why, and what would make them switch?

Face-to-face and telephone surveys, focus groups, in-depth interviews and mystery-customer programmes across regions, segments and languages.

Public opinion and polling

What do citizens actually think, and how is it changing?

National and regional polling. Every published poll carries its sample, mode, dates and margin of error.

Social and digital intelligence

What is the public saying about you right now, and should you act?

The public conversation captured, cleaned, coded and validated in Somali, English and Arabic. Brand health, crisis signals, competitor watch. Available now.

Data and analytics

What does the dataset prove, and what does it only suggest?

Weighting, segmentation, statistical testing and thematic coding, with limitations written into the deliverable rather than the footnotes.

Strategic advisory

Given the evidence, what should we do first?

Brand, communication and go-to-market recommendations sequenced into quick wins, operational fixes and strategic moves.

Monitoring, evaluation and learning

Did the programme work, for whom, and how do we know?

Baselines, endlines, KAP surveys, third-party monitoring and perception studies built for donor-grade accountability.

A standing read on Somalia, not a one-off report.

Fikir is turning its research capability into recurring intelligence: a quarterly national barometer, sector outlooks, a weekly business briefing and always-on social intelligence.

Explore Fikir Intelligence

WeeklySomalia Business Intelligence WeeklyPlanned QuarterlyFikir Barometer, the national opinion trackerIn design QuarterlySector Outlook briefsPlanned Always onSocial intelligence for brands and institutionsAvailable

Reading the public record.

Four of five published studies, each built by coding public conversation around a major Somali brand, by hand, in up to three languages. Together the five cover 3,456 comments. Each ends with what the organisation should do.

Telecoms. A mobile-money apology on Facebook. 625 of about 1,200 comments coded.

Only one in nine accepted the apology.

46% of coded comments were critical or mocking. 11% rejected the technical-fault explanation outright and read the outage as deliberate. Four in five critics tied it to a pattern or wanted their money back. The outage ended; the story it told did not.

It feels like you've been pulled into politics, and that doesn't serve you.Translated from Somali
How it was done
CaptureEvery public comment transcribed into a record.
CodeSentiment and theme, in Somali and English.
ValidateManual review of the major categories.
LimitSelf-selected commenters, not the customer base.
46%critical or mocking

Banking. A bank asked the public what it would invest in if financed. 2,001 comments in three languages.

The public confirmed three prompts and added one.

Of 1,556 people who named a sector, 41% said business and trade, 34% land, 23% farming. Livestock surfaced unprompted. Vehicle finance, which was prompted, underperformed. About one in twenty pushed back on the bank itself, and half of those feared it would take their idea.

So you're going to steal our ideas, then?Translated from Somali
How it was done
Capture2,001 comments transcribed, de-duplicated, cleaned.
CodeTagged to sector in Somali, English and Arabic.
ValidateStratified manual review.
LimitAn engaged public, not a representative sample.
41%chose business and trade

Telecoms. A bundle discount promoted to a price-sensitive market. 226 of about 451 comments coded.

Customers did the arithmetic in public.

17% argued the price had barely moved and showed their working. 12% benchmarked rivals in real time. Critical replies outnumbered positive ones two to one, and they cited exact prices. The offer was judged against the market, not in isolation.

Nothing actually got cheaper. Do you think people can't do the math?Translated from Somali
How it was done
CapturePublic comments transcribed into records.
CodeSentiment and theme, Somali and English.
ValidateManual review.
LimitHalf the thread coded; stated on the method page.
2:1critical to positive

Financial services. An $800,000 drought donation announced on Facebook. 447 of 529 comments coded.

Goodwill is earned by the gift. Trust is kept by proof.

41% praised or blessed it; 7% were critical or doubtful; the platform's Love reaction led. Underneath sat a sincere minority asking whether the help reached people, and 4% asking for wells rather than rice. The next post should show delivery, not announce it.

May God accept it from you, but we don't see anyone who actually received it.Translated from Somali
How it was done
Capture447 of 529 comments.
CodePraise, blessing, scepticism and constructive asks as separate themes.
ValidateManual review.
LimitSignal from an engaged public.
41%praised or blessed it

Subjects named are the brands whose public posts were analysed. The studies read the public record; they are not statements about commercial relationships. Public comments are a self-selected audience, and every method page says so.

All five studies and the research agenda

"Nothing actually got cheaper. Do you think people can't do the math?"

A customer, translated from Somali. One of 226 comments coded on a telecoms discount. The offer was judged against the market, not in isolation.

Local fluency. Published method. Built around your objective.

Somalia-groundedSomali-led and staffed, headquartered in Mogadishu, with fieldwork reach across the Federal Member States. We read the comment in the language it was written in and the market from inside it.
IndependentNo political campaign work and no advocacy messaging. The rule is written into contracts. We publish what the data shows, including when it is inconvenient.
Evidence-ledEvery published poll carries a methodology note. Every social study states how many comments were coded out of how many existed, and what that can and cannot tell you.
Objective-ledEngagements are designed backwards from the client's objective and end with a sequenced playbook, not a summary of findings.
International standardProcesses aligned to ISO 20252 for market, opinion and social research. Digital field collection on tablets. A formal data-protection and ethics policy.
Proprietary intelligenceRecurring products and original indices, from a quarterly national barometer to a trust and reputation index, so clients get a standing read rather than a snapshot.

Scope an engagement.

Two choices. We suggest the simplest design that can answer the question. The real scoping conversation follows.

Your sector
Your objective
Suggested design

Quantitative survey across the main urban centres, face-to-face on tablets, plus social intelligence in three languages to read price sensitivity against competitors.

Start a brief with this

Have a project in mind?

Brief us in about six minutes. Tell us your objective, who you need to hear from and when you need the answer. The Fikir team reads every brief and comes back with a design, a timeline and a cost.

  1. 1About you
  2. 2Your objective
  3. 3Scope
  4. 4Timing and budget
Start a project briefOr ask a quick question