Research and Investigation

Good research depends on good questions. But it needs a lot more than that: patience, perseverance, flexibility — and the ability to work at speed. This resource focuses on the whole process of investigative reporting. It follows a case study through its various stages, and suggests some strategies for doing good research on deadline.

Doing good research on deadline.

You work for a weekly investigative paper. Your editor calls you into his office. “I’ve just had a phone call,” he tells you. “An NGO in the rural areas says farm workers are being evicted by big agricultural companies, but are afraid to talk about it. The provincial government won’t do anything, they say. This sounds like our kind of story; we can put pressure on the authorities to stop the evictions. Here’s the name of our informant. Get on it. I want this for page 3 next week…”

You have a week. Where do you start? Take a few minutes to plan your strategy.

Make notes before you read on

Maybe you suggested doing some reading on land rights or the company involved. Both of these are good ideas – but an investigative strategy always needs to be based on:

  • A clear understanding of what needs doing
  • Balanced with a realistic recognition of the resources – including time – you have available

Realistically, you don’t have the time to become an expert on all aspects of land rights and problems. It can be very intimidating if you believe that a journalist needs to know everything. In fact, a good journalist needs a different skill: knowing how to find out anything. So you need to find people who are expert in land and land rights.

  • First, obviously, you should talk to the initial caller to get more information on the story. This person works for an organisation in the area — so you have to consider that he or she may be biased. Very few people ring up newspapers with stories unless they have ‘an axe to grind’.

Take a few minutes to think about the advantages and disadvantages of using biased sources:

Biased sources can be very useful. They often have detailed inside knowledge, and can give you strong questions to ask the opposing side. But you need to remember that the information comes from one side of the case; it will need cross-checking. And representatives of organisations are often speaking on behalf of much larger groups of people; when they summarise group views, they may organise and edit community opinions in ways which change or exclude important aspects — sometimes quite unintentionally.

So you need to think about what the bias might be and read, interview and research widely, to put the informant’s statements in context.

Having got more detail from the initial informant, you can use the notes of that conversation to start planning your first lines of investigation.

  • Next, you should hit the phones. The phone book is your most important reference book: if all else fails, work through every entry with the relevant surname or keyword until you find the right person. Ring every relevant source you can think of. If you can’t think of any, use the phone book (or, these days, the web) to trace organisations with “land” in their names.

Start with your personal contacts, or fellow journalists dealing with the relevant rural area, land, labour (farm workers), environment and agri-business. Among other sources you should phone are university departments and libraries (including your own paper’s archive). Work outward from the sources you know to identify the sources you don’t yet know.

Ask these contacts:

– Do you deal with /carry information on…?
– Who is the relevant person in your organisation to tell me about…?
– Do you know of anything (reports, articles, books) that’s been published about…?
– Can you suggest someone else it would be useful to talk to…?

Always take notes of these conversations. Get contact details to add to your contacts book for future use. Build up a “contacts tree” using the ‘working outwards’ strategy until you find the people you need.

  • Notice we haven’t mentioned government sources yet. You do, of course, need to get official views. But you need to be well-informed to ask useful questions, so this first stage of investigation may not be the best time to do a formal official interview. Also, you may waste too much time setting up an interview with someone who – when you’ve refined your research – may not even be the right person! If an official will talk to you off the record or informally, that’s best at this stage. You can always go back formally to get confirmation.


Sometimes either interviewer or interviewee makes the wrong assumptions about the status of ‘informal’ interviews, and this can lead to problems later. So if there is any possibility of misunderstanding, it’s best to ask: ‘Are we on or off the record here?’

  • “On the record” means you can use everything you’re told.
  • “Off the record” means you can use the information, but not in a way that allows the source to be identified.
  • “Background only” means don’t use this at all; it’s just to help you understand the context.

If you want to confirm off-the-record or background-only information, take it to someone else and (without revealing your source), ask them if such-and-such might be true and if they’ll confirm it on the record.

What next?

Your phone calls should have given you further names and contacts to follow, plus lists of reference material. This may include “grey” material — material which is widely circulated but which may not have been formally published (e.g. studies commissioned from private organisations, academic dissertations) or which may be officially confidential.

You have located eight possible phone contacts who may be prepared to talk to you. One of them sounds like the main expert on rural land rights in the area, but does not respond to messages. You feel you’re really stalled until she gets back to you.

From the web and your own paper’s archives, you have also found a dozen newspaper articles on farmworker evictions, two NGO case studies on landholdings in the rural area concerned, and a post-graduate dissertation on national land law reform. Tomorrow, you travel out to the countryside with a photographer. How should you spend today? Take a few minutes to note your answer before you read on:

You might feel that you should get through as much as you can of the reading. But this isn’t always the best strategy. Reading is slow and – especially if you’re doing internet research – you may have identified not a few, but a few hundred relevant references. You could easily spend the whole night reading and still not finish, or have time to make sense out of what you have read.

Your main choice is actually between reading or talking to contacts. So how should you divide your time between them? Time is precious.

Once you understand the issue better, much of the reading you’ve located may be irrelevant. At this early stage, skim-read only the broad, background stuff — in this case, probably the newspaper articles. Make sure your web-search has not been too wide (putting “…” around the key words will give you only those articles that include them all). Bookmark any web references that look interesting, so you don’t have to spend time later searching for them again.

Don’t waste time on that elusive expert. Chase up as many of the people on your phone list as you can. And go back to the people you first talked to with any related questions you’ve thought of.

At this stage, you need to search broad rather than deep.

  • Make a mini-timetable in advance, working backwards from your deadline. Decide how important each segment of the research is, and how much time you can afford to spend on it. Never waste extra time chasing up just one elusive person, document or figure — find another way to get what you need, or some other information that will do instead.
  • Plan your questions for the live and follow-up phone interviews based on what you are finding out. As new inputs land, decide in more detail what might be going on, what is being alleged, and who might be the culprits. You should be in a position to replace general, abstract terms like “corruption” or “injustice” with concrete, specific, allegations that X did or did not do Y to Z.
  • And begin putting yourself in the shoes of the people you are talking to. What might their agendas, motives, hopes and fears be? This way you can get a perspective on what they tell you.

You’ve skimmed the basic reading and had a number of very interesting phone interviews. And you’ve visited the district and seen the situation for yourself. You’ve found out the following:

  • The evictions are taking place only in one district. And the four families concerned have actually been very open in talking about their problems. In one community newspaper article, they blamed a district civil servant, Job Nkomo, for being lazy and not responding to their complaints. They repeated this when you spoke to them.
  • The commercial farming company concerned HAS been buying up land and trying to move small farmers off it. But it has been using legal process and following the rules. Your law expert says the company has not broken any rules. And elsewhere in the province, there are at least a dozen cases on record of evictions being successfully contested. The provincial government has been very energetic in publicising land rights, and there are posters up almost everywhere — although you didn’t see any in the district where the evictions are happening.
  • The NGO that initiated the complaint was only established three months ago and is still at the stage of securing funding for itself. Apart from establishing a small office, advocacy for the evicted families is the only activity it has undertaken so far. So maybe its agenda, while sincere, has elements of wanting to make a reputation for itself..?

Remembering your brief (“This sounds like our kind of story; we can put pressure on the authorities to stop the evictions”) can you now go ahead with the story? And what should your next steps be? Take a few minutes to decide on your answer before you read on:

  • Yes, there is still a story there. But it’s a different story: one about a small area which has been left behind by provincial land rights work — possibly because of a grossly negligent district official. The NGO’s agenda is to establish itself as a representative of the community, but its input is not very relevant; there are better quotes from the local people themselves. You can even see the new headline in your imagination: “The land the law forgot…”

Never be afraid to redefine your story in the light of new information!

Next steps:

  • Tell your editor what’s happening. She may need to re-plan space or place this different story on another page. You MUST do this as early as you can.
  • Now is the time to deepen your research and discard what’s irrelevant. Most of the reading can probably go back on the shelves, except anything dealing with that particular district. And your notes on the broad situation of the province may no longer be useful.

It hurts to discard work you’ve done. But you must. File your old notes: they may prove useful for some future story.

  • Now you can look for meaningful official comment. Face Mr Nkomo with the accusations against him. Ask the provincial (and/or national) land departments whether their policies should be applied everywhere, and how they feel about areas the new law hasn’t touched.
  • You will, however, also need to go back to the most interesting contacts and references to find out if there is anything in the history of that district which has led to it being so neglected, and to ask other relevant questions.
  • What you’re looking for is a sequence of events and/or chain of evidence that links the district official’s inaction to the evictions of the four families. It helps if you can find a rule that has been broken or an action that is required but was not taken. Again, you need concrete, specific evidence to replace the abstract accusations like ‘laziness’ that people have laid against Mr Nkomo. And be sure to deepen your understanding of Nkomo and his actions if he is to be your focus. You don’t want a defamation suit — nor do you want to omit any aspects of his misconduct.
  • Flag the risk of defamation with your editor. She may need to pass the finished story to a lawyer for advice.
  • Forget anything you can’t verify. Look for conflicting points in your notes — can they be reconciled? And what do your “biased” sources have to say about it all? Try to get on-record comments. Check, cross-check and check again.
  • It’s very rare to find absolute proof of something – the ‘smoking gun’, as it is called. That is what you are looking for: the documentary proof Nkomo did not respond to an appeal: for example, a memo from him saying the problem did not exist. But:
    • you may not have the time or resources to find this (Remember, this is now a smaller story of local government neglect, not a big story about multinational agribusiness. It is still worthwhile, but may not be accorded the same resources.)
    • finding it may require steps like searching an office that are illegal and/or ethically unacceptable
    • it may not exist!

You can build up almost as convincing a case by sheer weight of evidence: a detailed time-line, for example, of what happened, what should have been done, and Nkomo’s failure to communicate, supported at each stage by facts and relevant comments from role-players and experts.

Now you can start to write…

We can see here the basic stages of investigation:

Define the initial question
Build up contacts
Do broad research
Look again at the question. Don’t be afraid to redefine it
Discard irrelevant material
Do narrow, deep research (including back-tracking)
Plan and write the story

The two question-defining stages are crucial. A story doesn’t exist until a subject area has been given angle and focus. And if you’re failing to come up with answers, it may be because you’re asking inappropriate questions.

Researching on very short time

We’ve looked at the example of an extensive story done over a few days so that we can focus on every stage of the process in detail. But what happens when you are trying to research two stories in half a day before the deadline for a daily paper? Exactly the same principles hold good – that’s why we looked at a one-week example.

The temptation is to cut the planning time, and dive straight into the research. That’s a mistake. Good research – whatever your timeframe – depends on good planning. That’s particularly true when your time limits you to fewer sources and fewer questions – you have to make sure that you use your time wisely; asking the right questions of the right sources. When you’re working to a tight timeframe, the stages of the research don’t change:

  • Define what you’re looking for. Spend more time here, sorting out story priorities, so that if you only have time for a little research, it will be relevant. Check with your editor how long the story should be, and what angle the paper is seeking.
  • Research broad. You may have to settle for one Internet-sourced overview and contacts from two, rather than a dozen, phone calls. But it’s still vital that you check you have a story, and that it’s what you think it is.
  • Redefine. Does your angle still hold? Then start writing at this stage. Use the research narrow stage simply to confirm key facts. And remember two important points:
  • Some stories don’t need deep investigative research – make sure the story is worth the effort (and consult).They simply need accurate fact collection and clear writing. This is particularly true for those sections of the paper which still carry fairly short, “hard” stories. Even here, you need to make sure the story is valid and worthwhile, but having done that, you can go ahead and write it.
  • And some stories are simply too important and/or complex to be tackled in two hours. Visiting overseas journalists are often shocked by how fast writers here are expected to work, and how many stories they tackle in one day. The visitors believe this may encourage shallow journalism. So don’t be afraid to make your case to your editor if you believe your story needs more time and research. But don’t use this argument to cover up work not done. You’ll need to have gone through the early investigative stages and have solid facts and reasons if you want more time. And there may be scope for a compromise, where you deliver breaking facts on time, but plan for a longer background or context piece later in the week.


Your newspaper has limited space. But even if it did not, readers will not neccessarily have the patience to read all the (often dry) facts and figures you have collected. So bear in mind the following hints when you actually write the story:

  • Think about what will attract your readers, today. Put that at the top.
  • Use newsroom team-working and communication so that editors, sub-editors, photographers etc are all on-message about the story.
  • Look for character and narrative (a story). You might focus on one of the evicted families, or one family member, as a door to get into the explanation about land law rights. Your narrative shows the problem (a law that isn’t being enforced) and its resolution – either what could be done in this district, or what the district next-door is doing to avoid the problem.
  • Make the abstract concrete and the general specific. Don’t say “The official’s uncaring attitude has created havoc.” Say “Because Mr Nkomo gave the wrong advice, the Sitholes lost their farm.”
  • Show, don’t tell. Don’t editorialise about incompetent district officials. Show us what harm they do.
  • Explain everything. Use time-lines, fact-boxes and plain-language explanations of legal jargon so that your readers understand clearly what should have been done and was not.
  • If you have photographs, or are a broadcast journalist with access to picture and sound-clips, use those to tell the story too, not just as illustrations of people’s faces. How about two contrasting photos: the many posters about land rights in a neighbouring district office, and the blank walls in Mr Nkomo’s office? Or his coat on the back of his empty chair..?
  • If you’re writing for the Web, use updates, chat forums, links to related stories and resource sites, reader voting opportunities (‘Should Nkomo be fired? Vote here”) links to video of interviews etc to make the story more vivid and involving.


Newspapers that have a web presence or good phone links are beginning to draw readers into investigative work. They might report a story like the one above, and then invite readers to add to it. When the US Fort Myers News-Press did this on a story about over-billing for local utilities, they were astounded by the response. Experts among their readership sent them analyses of balance sheets and technical papers, and there were thousands of phone calls and e-mails. Read the full story on the newspaper’s website

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Forum for African Investigative Reporters