The reader intends to act

How-to guide writing services
for readers who will actually try this, at the moment they need it.

How-to guide writing services in Nigeria are judged by one test that nothing else on this site is judged by: does it work when followed by somebody who has never done it before. Not whether it is clear, sound or well written, though all three matter. Whether a stranger can follow it and end up where the guide said they would end up.

One reader, in one situation, with what they already know stated out loud. Steps in the order that reader does them. What done looks like, said before the first step. And a test on somebody who has never done it.

WHAT THE WORK IS 01 / 04

Every other kind of writing can be unfinished and still be useful. This one cannot, because the reader came to do the thing.

  • 01 ReaderOne, in one situation.
  • 02 StepsIn their order, not yours.
  • 03 The testOn somebody new.
Tested on a stranger Assumptions stated out loud Steps in the reader's order Done defined up front

The Writely difference

A how-to guide is the only genre whose reader intends to act, so the only test that matters is whether it works when followed by a stranger.

Clear, sound, well organised and well written all matter here. They are also all secondary, because a guide can score well on all four and still leave the reader unable to finish.

The reader intends to act

Everything else on this site has a reader with a reason to be careful. A letter has a recipient who will decide something. A proposal has a counterparty weighing it. Minutes have a body answerable for them. A blog post has somebody who may well leave. A how-to guide has a reader who arrived intending to do something, and who will find out within an hour whether it worked. It has to be complete, because an incomplete guide is not a partial success for this reader, it is a failure. It has to be unambiguous, because an ambiguity wastes time the reader has already spent. And somebody else's mistake ends up being yours, which happens nowhere else in professional writing. You can also explore content analysis writing.

What to assume is the hard decision

The most consequential judgement in the genre, and the one least often made on purpose. A guide that explains everything is unusable at the moment of use, because the reader is trying to do a thing rather than learn about it. A guide that assumes too much abandons them at step one. Both failures are common, and they are common because the line is drawn by whatever the author happens to know rather than by anything about the reader. The fix is unglamorous: state the reader in one sentence, including what they already know, and then cut every step that reader would not need. You can also explore research proposal writing.

Experts write bad guides

The people best qualified to write a how-to guide are usually the worst at it, and the reason is structural rather than careless. They no longer notice the steps. A person who has done a task two thousand times does not run the checks, does not realise the screen is a different version, does not know which of the clicks they always make without thinking. Writing from that position produces a document that is a list of destinations rather than a set of instructions, and the missing steps are exactly the ones a beginner trips over. You can also explore research paper writing.

The test nobody runs

There is one check that answers the only question this genre asks, and almost nobody performs it, because it is inconvenient and takes an hour. Give the guide to somebody who has never done the task, watch them do it, and write down every point where they stopped, looked something up, guessed, or gave up. That list is the entire review. Nothing else on this site can be tested this way, and nothing else on this site needs to be tested this way. A guide that has never been followed by a stranger is a draft.

Find what your guide is missing

Eight things a how-to guide has to have, and most drafts have four.

None of these are about clarity. They are about whether a stranger can finish, which is a different and much harder question.

Write for one reader in one situation

A guide written for everybody is a guide that is wrong for the person holding it. State the reader in one sentence, including what they already know, and then state the situation, because the same task differs enormously depending on what the reader is trying to achieve and what they are working with. Getting this wrong is usually invisible, since the author fills their own assumptions back in and reads something reasonable. It also decides the length. A reader who has done it before needs three steps and a list of the differences. A reader doing it first needs twelve steps and a definition of every term.

Steps in the reader's order, not the author's

The most common structural fault is ordering the steps the way the subject is organised rather than the way the work is done. Those are frequently different, and the reader will not be able to reconstruct your ordering. So write the steps as a sequence somebody could follow without understanding the subject, and check each one asks only for things already available at that point. A step that needs something introduced later is a step the reader will stop at.

Say what done looks like before step one

Most guides explain how to start and never say how to tell whether they finished, which leaves the reader with no way to know whether to stop, carry on, or start again. Put the outcome at the top, described concretely enough that the reader can check it against what they end up with. This is unglamorous and it is the difference between a guide that gets followed and one that gets abandoned, because a reader who cannot tell whether they are finished will assume they have not finished and keep going.

Name what can go wrong, and where

A guide that only describes the happy path is a guide that fails at the first surprise, and the reader discovers at that point that the document did not describe their situation. Say what commonly goes wrong, where in the sequence it goes wrong, and what to do about it. Keep that material near the step it belongs to rather than gathered at the end, since a reader who hits a problem is reading that step and not the appendix. Resist the temptation to cover every possible failure, because a guide that describes every exception is never finished and therefore never usable.

Cut the steps your reader does not need

Every unnecessary step is a place the reader can go wrong, and most guides carry several. Some exist because the author finds them interesting. Some exist because the platform used to require them. All of them burden the reader. So take the finished draft and go through it asking, for each step, whether this reader would skip it, and cut the ones they would. What is left is usually shorter and more reliable, which is the right trade, and the counter-intuitive result is that removing steps makes a guide more likely to be finished rather than less.

Get the material by watching, not by writing

A guide assembled from what the author knows contains every step they no longer notice, which is the defect the whole genre is prone to. The fix is procedural rather than creative: watch somebody competent do the task, take notes on the order, and specifically record the places where they paused, checked something, or asked which screen to use. Those pauses are the steps the author would have skipped. Do it more than once, because two people do the same task in a different order, and the difference between them is where the guide was ambiguous. Asking the person afterwards what they were unsure about is also useful, since they will have rationalised the uncertainty by the time you ask.

Pictures help only when they carry information

Screenshots are the most-used and least-examined element in this genre, and most of them add nothing, because the reader needs to recognise a screen rather than admire it. A picture earns its place when it shows something words would take a paragraph to say: where a control is on a crowded screen, what a finished result looks like, what a particular state means, or what the screen looks like on a different version. A picture showing a clean desktop does not transfer to the phone the reader is holding. Cover the text and look at the picture. If nothing is lost, it was not carrying anything.

What this service cannot do

Four limits, worth stating before the brief rather than after it. We cannot run the test for you, because it needs somebody who has never done the task and that person has to be found and watched rather than simulated. We cannot write steps for a procedure we have not seen performed, since that is exactly the failure the genre is prone to. We cannot promise that a guide will remain current, because procedures change underneath documents and nobody controls that. And we cannot write a guide for a reader who does not yet exist, which in practice means the reader has to be settled before the drafting starts rather than decided afterwards.

Never been followed by a stranger? It is a draft. One hour watching somebody attempt it tells you more than a week of revision.

A better brief, a better guide

Send us who the reader is, what they already know, and a recording of somebody doing the task. The guide follows from those three.

The useful help begins with the reader rather than with the procedure, because the procedure is usually known and the reader is the thing that decides what has to be written. This service writes how-to guides as instructions that get followed: one reader, assumptions stated, steps in the reader's order, what done looks like said before step one, and the failure points named where they belong. It will not write steps for a procedure nobody has watched being done.

  • Say who the reader is, in one sentence.
  • Say what they already know and what they have not done.
  • Send a recording of somebody doing the task.
  • Say what they will have at the start.
  • Tell us what done should look like.
  • Name anyone who can test it afterwards.
Share your brief
GUIDE CHECK Before you publish

Six things before you publish.

One reader, stated up front
What they know, stated plainly
Steps in the reader's order
Done described before step one
Failure points named in place
Followed by a first-timer

How the work is done

From who is holding this to a guide that has been followed by somebody who never had.

Reader, steps, exceptions, test. The order is not a preference. A guide drafted before the reader is settled ends up written for its author, which is the failure the genre is known for.

01 / THE READER

Settle who is holding the guide

Write the reader in one sentence, and include what they already know, because that clause is what decides the length and most guides get it wrong by describing a person rather than their knowledge. Then settle the situation, since the same task differs enormously depending on what the reader is trying to achieve and what they have in front of them. And check whether this is genuinely one reader, because if the beginner and the experienced reader are both real, one guide becomes two, and a guide written for the average of them suits neither. You can also explore research proposal writing.

02 / THE STEPS

Watch somebody do it, then write in their order

Do not write the steps from what you know. Watch somebody competent do the task and note the order, and specifically record the places where they paused, checked something, or asked which screen to use, because those pauses are the steps the author would have skipped. Then write it as a sequence somebody could follow without understanding the subject, and check every step asks only for things already available at that point. You can also explore essay writing.

03 / THE EXCEPTIONS

Name what goes wrong, and where

Now add the failure points, because a guide that describes only the happy path fails at the first surprise and the reader discovers at that point that the document was not written for their situation. For each common problem, say where in the sequence it tends to happen and what to do about it, and keep it next to the step it belongs to rather than gathered at the end, since a reader who hits a problem is reading that step and not the appendix. Then describe what done looks like, at the top, not the end.

04 / THE TEST

Hand it to somebody who has never done it

The last step is the one almost nobody performs, and it is the only one that answers the question the genre asks. Find somebody who has never done the task, give them the guide and the starting materials, and watch. Write down every point where they stopped, looked something up, guessed at something, or gave up, because that list is the entire review and nothing else comes close to it. It finds the skipped assumption, the unexplained term, the screen described from memory, and the place where the guide asked the reader to decide something. Then fix those and repeat once, because the second attempt surfaces a smaller set. A guide that has never been followed is a draft, whatever else is true of it. You can also explore academic editing and proofreading.

What actually gets handed over

Four documents. The guide is the one that gets published and the only one a reader has ever seen.

The guide is the deliverable. The three before it are what make it finishable, and they are the part most senders never produce at all.

The reader note

One page naming the reader in a single sentence, including what they already know and what they have never done, plus the situation they are in and what they will have available when they start. It looks like a short brief and it is the document that decides the length, because a reader who has done the task needs three steps and a beginner needs twelve. Short, and the reason a guide stops growing by accident. It is also the document that shows a second guide for the other reader is needed rather than a longer one.

The step list

The sequence, written as it was observed rather than as the subject is organised, with a note against each step on where it came from and what it assumes is already in place. The list arrives having been assembled by watching, not by writing from memory, which is the whole difference between a set of instructions and a list of destinations. It is also what makes the guide revisable, since a later editor can see which steps were observed and which were assumed.

The failure note

The common problems, where in the sequence each tends to happen, and what the reader should do about it, kept next to the step it belongs to rather than gathered at the end. Placement matters more than coverage here, because a reader who has hit a problem is reading that step and not an appendix, so a well-organised failure section in the wrong place does not get read at the moment it is needed. What belongs here is the handful of failures that actually happen, in priority order, because a reader who has just gone wrong needs the commonest one first.

The guide and the test record

The guide itself, plus the record of the last time a first-timer followed it. That record is a short document and it is the most valuable thing in the set, because it is the only part of the writing process in this genre that measures the thing the genre is for. Every point where somebody stopped, looked something up, guessed, or gave up, with what they were doing at the time. A guide shipped without one is a draft that has been published, which is a state most guides are in.

The test record is the part senders skip and the part that measures the only thing this genre is for. One hour watching somebody new tells you more than a week of revision.

What a guide has to settle

Six things, and most published guides settle two.

These are the acceptance criteria, and each one is something you can check in five minutes before the guide goes out rather than discovering when somebody cannot finish it.

  • One named reader, with what they know stated.
  • What done looks like, said before the first step.
  • Steps in the order that reader performs them.
  • Every step asks only for what is already available.
  • Common failures named next to the step they happen at.
  • A first-timer has attempted it, and the notes were kept.
THE GUIDE Six things

Six things a guide must settle.

One reader, named and known
Done described before step one
Steps in the reader's order
No step needs a later step
Failures named in place
A first-timer attempted it

Explore related academic support

The evidence a guide's steps have to rest on.

A guide that states what something is, how well something works, or why a method holds up is making an evidential claim, and those claims are what the steps are standing on. These are the pages that produce that material and check it before it becomes an instruction.

A guide that states a fact or a figure? Say so in the brief. An instruction is only as trustworthy as the claim it rests on, and a wrong fact inside a procedure gets followed by more people than a wrong fact in an article.

Where the standards come from

First: what kind of document this is. Then: how the ones that work are written.

The first resource here is more important than the other seven combined, because most guides are written by people who have never settled whether they are writing a tutorial, a reference entry or an explanation, and that ambiguity shows up as a document that is none of them. Every resource here was confirmed to answer before it was listed. These external resources are informative references, not Writely service pages.

WHAT KIND OF DOCUMENT THIS IS

Diátaxis

The framework that splits four kinds of document which are constantly confused: a tutorial for a beginner learning, a how-to guide for a competent person doing a task, a reference for somebody looking something up, and an explanation for somebody who wants to understand. It is the sharpest thing available on this page because the confusion has real consequences. A how-to guide that has drifted towards being a tutorial pads the steps for a reader who does not need them. One that has drifted towards being a reference omits the steps a reader needs entirely. Knowing which of the four you are writing tells you what to cut, which is the hardest decision in this genre.

Open resource
HOW TO WRITE STEPS THAT GET FOLLOWED

Google Developer Documentation Style Guide

A published programme's own rules for procedural documentation, and the most detailed public set available on the specific mechanics of a step. Its material on the second person, on starting a step with a verb, and on describing a screen rather than referring to it are the parts worth reading first, because those three are where most guides fail for reasons that have nothing to do with clarity. Also useful is what it says about time estimates and prerequisites, both of which readers rely on heavily and both of which are usually guessed. Written for a technical audience but the mechanics transfer unchanged to a process anybody can do, which is why it appears here rather than only on technical pages.

Open resource
PLAIN WORDING AND ACCESSIBLE STEPS

US Web Design System

A national design system, and the two parts of it that bear on this genre are its plain wording guidance and its rules for content structure. The wording work is the familiar material on front-loading and using the reader's words, but it is applied to procedures rather than to pages, which changes the recommendations: an instruction cannot be made scannable by formatting alone, because the reader needs the action not the appearance. The content structure rules are useful for a different reason, which is that they force headings to describe what the section does rather than what it is about, which is exactly the fix for a guide a reader cannot navigate while doing the task.

Open resource
A SERVICE STANDARD ON INSTRUCTIONS

Writing for GOV.UK

The content design guidance written for public bodies, which is the closest thing to a working standard for instructions written by people who did not choose the genre. Its principles are few and they hold: put the answer first, use short sentences, use the reader's vocabulary rather than the department's, and cut anything that does not help somebody do the thing. What makes it worth a page about how-to guides is the sheer volume of procedure it has had to write under those constraints, with readers who are often stressed, often unfamiliar with the organisation, and often doing it once. The revisions that worked are a better guide to this craft than most books about it.

Open resource
LARGE-GUIDE PRACTICE AT SCALE

GitHub Documentation

A documentation set large enough to have hit every problem in this genre and to have recorded what it did about them. Read it as practice rather than as theory, which means looking at how a procedure is structured when it has twenty variants and how the common path is kept apart from the rest. Two things are directly useful. Their handling of what to do when the reader's platform differs, which is a problem almost every software guide has and almost none solves well. And their decision to keep short tasks short rather than padding them to a house length, which is the exact opposite of what most organisations default to and is right for the reader's sake.

Open resource
WHERE READERS STOP AND WHY

Nielsen Norman Group

The research on how people behave while following instructions, and the source for the claim this page rests on, which is that a reader deciding to continue is deciding something other than whether the writing is any good. What they are deciding is whether they are in the right place and whether they are making progress. That is why a guide fails in the first screen far more often than in the last, and why a step that quietly assumes something is worse than one that is merely vague. Their work on abandonment during task completion is the evidential backing for testing with a first-timer rather than reading the draft again yourself.

Open resource
HOW PLAIN IT HAS TO BE

PlainLanguage.gov

The government's plain writing portal, and the corrective to the belief that plain writing means simpler vocabulary, since it mostly does not. For this genre it means something slightly different from usual, because the reader is usually working from a screen while doing something else, which is the worst possible conditions for reading. So the techniques that matter most here are the ones about structure: one action per step, the action at the start of the sentence, and no step requiring a reader to hold an earlier one in their head. The checklists are short enough to apply to a draft rather than merely admire, which is what makes them useful on a working guide.

Open resource
WHAT AGEING LOOKS LIKE

MDN Web Docs

The technical reference set that is the best available showing of the failure this genre cannot avoid, which is guides going out of date without anybody noticing. Its browsers-compatibility material, its deprecation notices and its stated support policies are all ways of making the age of a document visible on its face, which is the only practical defence against a procedure that quietly stopped being true. Read it for those three habits rather than for the technology. The practical question for any guide is when it was last checked by somebody who performed the task, and most guides have no answer to it at all.

Open resource

Local context, wider reach

Nigeria at the centre. Africa in view.

Writely is rooted in Nigeria and designed for remote academic and business support. Whether you are documenting a process for a company, a university, a polytechnic, a college of education, a research institute, a hospital, a non-profit, a church, a campus, a home office, or from another part of Africa or another country entirely, you can work on it here.

Remote-friendly academic support

How-to guide writing across Nigeria

Explore how-to guide writing for companies, teams and organisations in major Nigerian cities, communities and states. Four things are genuinely different here, and all four belong in the guide rather than in a later revision. Many of the readers will be doing the task on a phone with a mobile connection, so a guide that relies on pictures to convey anything will arrive almost empty and the steps will have to carry it alone. Power supply and connectivity interrupt procedures here in ways they do not in many other markets, and a guide that assumes an uninterrupted session is a guide that will lose readers at step three. And because instructions are frequently written once by whoever is nearest to the work, the discipline of watching somebody else perform the task is the part most worth outsourcing.

Lagos Abuja Port Harcourt Kano Ibadan Enugu Benin City Uyo Kaduna Calabar Jos Abeokuta Awka Akure Warri Owerri Zaria Sokoto Bauchi Maiduguri

Nigeria states and the Federal Capital Territory

Writely’s academic support is available remotely across Nigeria, including:

  • Abia
  • Adamawa
  • Akwa Ibom
  • Anambra
  • Bauchi
  • Bayelsa
  • Benue
  • Borno
  • Cross River
  • Delta
  • Ebonyi
  • Edo
  • Ekiti
  • Enugu
  • Gombe
  • Imo
  • Jigawa
  • Kaduna
  • Kano
  • Katsina
  • Kebbi
  • Kogi
  • Kwara
  • Lagos
  • Nasarawa
  • Niger
  • Ogun
  • Ondo
  • Osun
  • Oyo
  • Plateau
  • Rivers
  • Sokoto
  • Taraba
  • Yobe
  • Zamfara
  • Federal Capital Territory

Remote support across Africa

Country-level discovery

Writely welcomes conversations with students, researchers and candidates across Africa, including learners in:

Nigeria Algeria Angola Benin Botswana Burkina Faso Burundi Cabo Verde Cameroon Central African Republic Chad Comoros Côte d’Ivory Democratic Republic of the Congo Djibouti Egypt Equatorial Guinea Eritrea Eswatini Ethiopia Gabon Gambia Ghana Guinea Guinea-Bissau Kenya Lesotho Liberia Libya Madagascar Malawi Mali Mauritania Mauritius Morocco Mozambique Namibia Niger Rwanda São Tomé and Príncipe Senegal Seychelles Sierra Leone Somalia South Africa South Sudan Sudan Tanzania Togo Tunisia Uganda Western Sahara Zambia Zimbabwe

Questions, answered

Start with what you need to move forward.

If your question is not covered here, send the essentials in a chat. A precise brief usually leads to a more useful starting point.

Ask your question
What is how-to guide writing, and what is it not?
It is writing instructions that somebody else can follow to completion, which is a different task from explaining a process or describing how something works. The distinguishing feature is that the reader intends to act, and will find out within an hour whether the document worked, which makes this the only genre on this site where another person's mistake belongs to the author. It is not a tutorial, which teaches a beginner, and not a reference entry, which is looked up. A how-to guide is for somebody who could already do the task and needs the steps. Blurring those three is what produces a document that is padded for a beginner and incomplete for a competent reader.
What should a how-to guide contain?
Six things, and the order matters. Who the reader is, including what they already know. What they need to have ready before starting. What done will look like, described before the first step rather than after the last. The steps, in the order that reader performs them, each asking only for things already available at that point. What commonly goes wrong, next to the step where it happens. And a statement of when the guide was last checked. What it should not contain is background about the organisation or about why the process matters, because the reader came to do a thing and will leave when the thing is done.
How long should a how-to guide be?
Exactly as long as this reader needs and no longer, which is the honest answer and usually much shorter than expected. A reader who has done the task before needs three steps and a list of the differences. A first-timer needs twelve steps and every term defined. Both readers are real and they are not the same reader, which is usually why one guide cannot serve both. The commonest waste is not padding but irrelevance: steps carried over from a longer version of the process that this particular reader would never take. Cutting those steps makes a guide more likely to be finished, not less.
How many steps should a how-to guide have?
Fewer than the author expects, and the number matters less than whether the reader can hold the sequence in their head while working. As a rough guide, anything past about seven or eight becomes a staged procedure, which is a different document with stage checkpoints and a clear outcome for each. The practical test is that every step should have an outcome the reader can see. If two consecutive steps have the same outcome, they are one step. If a step has no visible outcome, the reader cannot tell whether they have finished it or started it, which is where most guides quietly lose people.
How detailed should a how-to guide be?
Detailed enough that a stranger reaches the outcome without asking a question, and no more detailed than that. Over-specifying is a real failure and not only a stylistic one, because every unnecessary step is a place the reader can go wrong, and a reader who follows a step that turns out to be unnecessary will usually conclude that something is wrong with the guide or with themselves. The specific things worth specifying are the ones where readers reliably get it wrong: where to click, what the finished state looks like, and what to do if the screen is not what the guide says.
How do I write steps that are easy to follow?
One action per step, with the action at the start of the sentence in the second person, so the reader can tell what they are being asked to do before they finish reading it. Describe the screen or the object rather than referring to it by a name the author happens to know. Never require a step to use something introduced later, which is the defect that most often loses a reader halfway through. And check the whole sequence by reading only the step headings, because for somebody working from a guide that list may be all they have.
What should a guide say about mistakes and errors?
The common failures, in priority order, each placed next to the step where it tends to happen rather than gathered at the end, because a reader who has just gone wrong is reading that step and not an appendix. Say what the failure looks like as well as what to do about it, since a reader often does not recognise it without being told. And resist covering everything, because a guide describing every exception is never finished and therefore never usable. The handful that actually happen are worth more than the full list nobody reads.
Should a how-to guide include pictures?
Only when they carry information the words would take a paragraph to convey: where a control sits on a crowded screen, what a finished result looks like, what a particular state means, or how a screen differs on another version. A picture of a clean desktop does not transfer to the phone the reader is holding, and a picture showing something obvious is taking up the reader's time. The test is simple and it is quick. Cover the text and look at the picture. If nothing is lost, it was carrying nothing and should be cut.
How do I know whether a how-to guide works?
By giving it to somebody who has never done the task and watching. Record every point where they stopped, looked something up, guessed, or gave up, because that list is the entire review and nothing else comes close. It is inconvenient and takes about an hour, which is why most guides are never tested and most published guides are drafts. Watch for three specific things. A step that asks the reader to decide something. A term used without being explained. And a screen described from memory. Repeat once after fixing, since the second attempt surfaces a smaller set.
How do I keep a how-to guide up to date?
Put the last-checked date on the guide and treat it as a real commitment, because that single line is what tells a reader whether to trust it. Set a review trigger on the things that change the procedure rather than on a schedule: a software update, a change of interface, a new person taking the role, or a failure somebody reported. And have somebody who was not involved perform the task on the review, because the author cannot see the gaps in their own process any more. Most stale guides are not neglected, they are simply not connected to anything that changes.
Can you write the guide or do I keep it?
We write it and you own it, because the reader note and the decision about what to assume are judgements only you can make about your own readers. What we produce is the reader note, the step list, the failure note, and the guide itself, along with a test record once somebody has attempted it. What we will not do is write steps for a procedure nobody has watched being performed, because that is exactly the failure the genre is prone to, and we cannot run the test ourselves. Where the honest finding is that two audiences need two guides, that is said before drafting rather than after.
Where is Writely available, and how do I start?
Writely is based in Nigeria and works remotely with companies, teams and organisations across Nigeria and the rest of Africa. Use the Chat with us button to say who the reader is and what they already know, because the reader note is written first and everything else follows from it. Send a recording of somebody doing the task, since that is the material the steps are built from, and tell us what they will have available at the start. Say what done should look like, and name anyone who could attempt the draft afterwards. From that we can usually say within one exchange whether this needs one guide or two.

Your next move

Send us a recording of somebody doing the task. We’ll write the guide from what they pause on.

Whether you need the reader settled, the steps sequenced, the failure points named or the whole thing tested on somebody new, start a clear conversation with Writely.

Request services