Your guide to content strategy frameworks

A content strategy framework is the structure you use to write a content strategy. It sets out the order of the work: understand the problem, decide the approach, then decide what you will do about it.

This is the framework I use, with the questions I ask at each stage. Copy it and change it to fit your organisation.

I would rather you did not follow it too closely. A framework someone else designed carries someone else's constraints, and the parts you cut are usually the parts that tell you most about your own situation. Templates are worse again: they hand you a shape to fill in, which is how you end up with a strategy that reads well and describes nobody's business. If you take one thing from this, take the order of the questions and throw away my answers.

What a content strategy framework is

A framework is the structure of a strategy. It sets out the steps and the order, and leaves the content of each step to you. That is the difference between a framework and a template: a template gives you the answers to fill in, a framework gives you the questions to answer.

A strategy identifies a challenge, sets out an approach to it, and commits to a set of actions. If it does not name the actions, it is a statement of intent. Vision statements, goals, objectives and values are inputs to a strategy. None of them is a strategy on its own.

A content strategy applies that to content. It governs what gets made, for whom, by whom, on what schedule, and how anyone will know whether it worked. It covers the parts nobody puts in a deck, including who owns which page, what happens when someone leaves, and what gets retired.

How content strategy and content marketing differ

These get used interchangeably and they are not the same thing, which matters because the confusion is expensive.

Content marketing is one use of content: attracting an audience and keeping it, in the hope that some of it eventually buys. It is a marketing activity with a marketing objective.

A content strategy governs all of your content, including the parts marketing never touches. Product copy, support articles, error messages, navigation labels, policy pages, the wording of a form. It decides what gets made, who owns it, how it is kept accurate, how anyone knows it worked, and what gets removed. Content marketing sits inside that, as one of the things the strategy resources and measures.

Most teams I meet are doing content marketing without a content strategy. The symptom is consistent: a publishing schedule that never stops, no agreed owner for anything already published, and no mechanism for deciding what to retire. They are producing volume because volume is the only decision the setup can make.

The practical test is what happens to a page 18 months after it goes live. If the answer is that nobody knows, you have content marketing and no content strategy.

Whether your business needs one

If you publish anything regularly, you are already making these decisions. The question is whether you are making them deliberately or one piece at a time.

The cost of not writing it down shows up in specific ways. Two people commission overlapping work. Nobody can say which page is authoritative on a subject. Content is published and never reviewed again. A person leaves and takes with them the only understanding of why a section exists.

A framework will not fix a resourcing problem. If you have one person and no budget, writing a strategy will tell you what you cannot do, which is worth knowing but is not the same as capacity.

The three parts

The structure below has three stages: diagnosis, guiding policy, and the actions that follow from them.

• Diagnosis: what is wrong, stated as a problem rather than as a goal.
• Guiding policy: the approach you will take, which directs decisions without specifying every one.
• Action: what you will do, who does it, and how you will know.

Almost every weak content strategy I have read fails at the first stage. It begins with a target, works backwards to a list of activities, and never establishes what the obstacle was. That is why so many of them are interchangeable: if you skip the diagnosis, the only thing left to write about is the activities, and everybody's activities look the same.

Diagnosis

The diagnosis is research, and it is where most of the work is. It answers three questions: what is getting in the way, what does that cost, and what would have to change.

Start from the business objective

Do not start by deciding the content strategy should increase revenue by a set percentage. Start by asking what the organisation is trying to do and where content is currently obstructing it.

Content work can reduce support contacts, improve conversion, shorten sales cycles and reduce the cost of producing the same thing twice. What it does underneath all of those is connect people to information that helps them decide. Most of the measurable effects follow from that.

Audience research

You cannot decide what to publish without knowing who it is for and what they are trying to do. The questions worth answering:

• What are your customers trying to achieve, and what stops them
• What do they need to know before they can decide
• What do the sales and support teams get asked repeatedly
• Who is the customer you want more of, and how do they differ from the ones you have
• Which segments are different enough to need different content

How to understand your audience and make sure your message resonates covers how to match a research method to each of those questions, and what to record.

Run a content audit

An audit tells you what you already have and what it is doing. If you have an established site, I would do this before deciding anything new, because a good deal of what you are planning to make already exists in some form. It is the least popular part of the work and the part that changes my mind most often.

Record for every page:

• What it is for, and which audience it serves
• When it was last reviewed, and by whom
• Whether it is accurate now
• How people arrive at it
• Whether anyone has visited it in the last 6 months
• What it asks the reader to do next
• Who owns it

Then look across the set:

• Which subjects have several overlapping pages, and which have none
• Which pages exist because a team wanted them rather than because a reader did
• Which content types are produced out of habit
• Whether the same claim is stated differently in different places

How people find the content is a separate pass, and it usually contradicts the first one:

• Which queries bring people to the site, and which pages they land on
• What the site search log shows people looking for and not finding
• Where the navigation labels do not match the words people use
• Which pages rank for something you did not intend
• Whether traffic can be separated into organic, paid, referral and direct, and whether anyone acts on the difference

Then the constraints, which decide what is realistic:

• What the content management system makes difficult or impossible
• Which templates exist, and whether a new content type needs development work
• Who has publishing access, and whether the number is too high or too low
• Whether pages can be redirected and retired cleanly
• Whether the site meets its accessibility obligations, since remediation may take priority over new work

The audit usually produces more retirement decisions than creation decisions. That is the useful outcome, and the one most audits avoid acting on. A page that is inaccurate and unvisited costs more to keep than to remove, because someone will eventually find it and believe it.

Internal challenges

The obstacles are often internal. Reviewing the purpose and priorities of the organisation tends to surface the gap between what is said and what is resourced.

• What is the organisation for, and who decides that
• What are the priorities this year, in order
• Who approves published content, and how long does that take
• Where does content work currently stall, and at which step
• Who owns the subjects nobody has claimed
• What capacity exists, measured in days per month
• What happened to the last plan of this kind, and why
• How will stakeholders be managed

These questions are close to fluff if answered in the abstract. They become useful when answered with names, dates and durations.

Why a broad objective hides the work

Take a goal like "revenue from the website will increase by 15% within 12 months". It sounds like a strategy and specifies nothing.

There are several routes to it. Increase conversion. Increase traffic and hold conversion steady. Increase the value of each conversion. Reduce the number of people who leave partway through. Each route implies different work, different owners and different costs, and some of them are not content problems at all.

The questions the goal does not answer are the diagnosis: how many steps are there before a conversion, which of them loses most people, are all conversions worth the same, who owns paid traffic, and what would have to be true for content to move any of it. Working through those turns a target into a set of identified obstacles, which is what you can plan against.

SWOT analysis

I include a SWOT with some reluctance, because most of the ones I am shown are generic enough to belong to any organisation in the sector. It earns its place only if every line is specific enough that a competitor could not have written it.

• Strengths: what you have that a competitor would need time or money to match
• Weaknesses: what consistently slows the work, stated as something you could change
• Opportunities: which of these you could act on within the year
• Threats: what would make this strategy the wrong one

Guiding policy

The guiding policy is the approach you take, based on what the diagnosis found. It sets the direction and rules out options without prescribing every decision.

Write it as a response to each identified challenge:

• The challenge, stated as the diagnosis found it
• The approach you will take to it
• Why that approach rather than the alternatives
• What would tell you it is not working

The last point is the one usually left out, and it is what separates a policy from a preference. A policy you cannot imagine abandoning is not a decision you have made.

A worked example. The diagnosis finds that support handles the same 12 questions repeatedly, that the answers exist but are spread across 30 pages, and that no page is authoritative. The guiding policy might be to hold one authoritative page per question, owned by the team that answers it, with everything else pointing at it rather than restating it. That rules things out: it rules out publishing a new page each time a question resurfaces, and it rules out marketing rewriting an answer without the owning team. The sign it is not working would be the same question arriving through support after the page exists, which would tell you the answer was hard to find.

Keep it short enough that people remember it. A guiding policy nobody can recall in a meeting will not shape what happens in that meeting.

Action

This is the operational half: what gets done, by whom, and on what cadence.

Decide who holds which responsibility. Web content roles, and which one your problem needs covers how the roles differ and which one to fill first. Content published without an owner becomes content nobody reviews.

At an operational level you are likely to need:

• Brand guidelines and a style guide, so the work is consistent without every piece being negotiated
• A content plan covering what is being made and why
• An editorial schedule built on the capacity you have
• The channels you have decided to use, and the ones you have decided not to
• A review cycle, with named owners and dates
• A content design process
• The tools, and a clear answer on which system holds the authoritative version of each thing
• Measurement, covered in how to set KPIs and marketing objectives

Where the problem is that the machinery does not run rather than that the direction is wrong, what marketing operations management is for covers the boundary between the two.

The actions have to support each other. A publishing schedule that assumes more capacity than the review cycle allows will produce a backlog of unreviewed content, and both decisions will look reasonable on their own.

Write the diagnosis first

Before planning any content, write a single page stating the three or four things currently getting in the way, what each one costs, and how you know. Use the audit and the internal questions above to fill it.

If you cannot complete that page, the research is the work, and any plan written before it will be a list of activities with a target attached.

And when you do write the strategy, make it yours. I have given you my questions because they are the ones I have found worth asking, not because the order is sacred.