Key Takeaways
- Start with one business outcome and a defined customer problem.
- Choose channels using audience evidence and your ability to support them.
- Plan proof, destinations and approvals alongside creative ideas.
- Review qualified enquiries and delivery effort, not just followers.
A social media strategy is a set of decisions about who you want to reach, why they would care and what the business can reliably deliver. A spreadsheet of scheduled posts is the operating calendar that follows those decisions. Starting with the calendar often produces activity without a clear reason for doing it.
Use these eight steps to prepare a small-business plan before you commit to more content or a monthly agency package. Write the answers down, assign an owner and agree when to review them. You do not need an elaborate presentation, but the choices need to be specific enough for someone else to act on.
Step 1: Choose a business outcome
Decide what the activity should support: relevant enquiries, repeat purchases, event registrations, recruitment or customer support. These are different objectives and need different content. Avoid a plan that promises to achieve all of them equally with a small budget.
Describe an outcome in terms the business can verify. For a consultancy, this could be conversations about a particular service from organisations that fit its delivery capacity. A like or a video view can help diagnose distribution, but neither proves that the person needs the service or has become a customer.
Write down the present position before choosing a target. If you do not know where enquiries come from, improving that record may be the first task. Agree who will judge lead quality and how they will feed the information back. Do not choose a target purely because it looks impressive in a proposal.
Step 2: Define the customer and the decision they face
Use actual sales questions, customer interviews and service records. Identify the circumstances that make someone look for help, the alternatives they compare and the information that makes them hesitate. Demographic labels alone rarely explain a buying decision.
An illustrative example: a growing business may know its website is difficult to edit but be unsure whether it needs a new platform. Its useful content might compare improvement and replacement, show an editing demonstration and explain migration risk. This is a planning example, not a claim about a measured client campaign.
Write a short audience statement: who they are, what has changed, what they need to decide and what proof they need. Include people who are not a good fit. Clear exclusions prevent the team spending time producing content for an audience the business cannot serve.
Step 3: Select channels with a reason to be there
Look at where relevant customers already ask questions, follow peers or discover products. Combine customer feedback with your existing channel data. Do not assume a platform is suitable because it is fashionable or because a competitor has a profile there.
Consider production and response capacity. A channel that needs regular demonstrations may be difficult if you have no permission to film the work. A channel generating direct messages needs someone able to answer them. The operating cost includes those responsibilities, not just pressing publish.
Choose a primary channel and a limited supporting role for others. Keep a note of why each one is included and what would make you reconsider. Our social trends guide offers test ideas without treating new features as mandatory investments.
Step 4: Build content around questions and proof
Create a small set of themes: understanding the problem, comparing approaches, seeing the work, preparing to buy and using the service successfully. For each theme, list questions your audience genuinely asks and the source material you can use to answer them.
Separate proof from illustration. A client case study needs accurate scope and permission; a numerical result needs a measurement period and source. If you are showing a proposed workflow or invented example, label it clearly. Do not turn an idea into a past client success story.
For instance, the Cornwall Flooring & Home Co. identity project can support a discussion of applying a visual identity across materials. It does not establish a sales increase. Use the brand-storytelling worksheet to turn real decisions into understandable content.
Step 5: Match each message to a next step
A person discovering a problem may need a guide; someone comparing suppliers may need examples, scope and pricing context. Someone ready to enquire needs a clear contact route and an explanation of what happens afterwards. Match the destination to that stage.
Audit the linked pages before building the campaign. Check the page on a phone, confirm that the offer is current and submit a controlled test enquiry. Verify that it reaches the right place. A confusing destination can waste effort even when the social content is relevant.
Keep promises consistent. If a post advertises a planning call, the destination should explain the call rather than demanding an unrelated purchase. If a guide is free, do not introduce an unexpected payment or an unclear marketing opt-in. The landing-page guide covers this handover in detail.
Step 6: Set a realistic budget and approval process
Budget for research, writing, design or filming, approvals, publishing, replies and measurement. Keep paid distribution separate from production so you can see where the money goes. Account for the business owner's review time and the person who supplies project facts.
Choose a rhythm you can sustain with those resources. A smaller number of useful posts can be a better starting point than an ambitious schedule that requires unreviewed filler. Agree how urgent changes are handled and who can pause scheduled content if circumstances change.
Create a pre-publication checklist for factual claims, image rights, customer permissions, legibility, captions and advertising disclosure. Use the ASA's brand-owned and paid-social guidance when planning commercial posts. Approval should mean a named person checked the final version, not simply that a draft reached a shared folder.
Step 7: Decide how you will measure and learn
Choose a business measure and a few diagnostic measures. For example, qualified enquiries show commercial progress, while link clicks and completed form steps may help explain where people stop. Avoid dashboards containing dozens of numbers that never change a decision.
Use consistent tags on external campaign links. Google's campaign URL guidance explains the source, medium and campaign parameters. Keep internal website links untagged so movement between your own pages does not introduce misleading campaign information.
Review the records with the people handling enquiries. Note spam, unsuitable requests and repeat customers separately. Consent choices and later visits can limit attribution, so do not assume every unassigned enquiry came from social media. Acknowledge the unknowns while improving the evidence you can collect legitimately.
Step 8: Agree a review date and decision rules
Choose a review period that matches your sales cycle and available volume. State what would justify continuing, changing or stopping an activity. Low traffic may support a practical qualitative decision without supporting a statistically reliable claim of uplift.
At the review, examine what was actually delivered, how much effort it required, what customers asked and what happened to the enquiries. Look for recurring misunderstandings or objections that the next content batch could answer. Do not change the entire strategy because one post had unusually low reach.
Keep a dated decision log. Record the change and its reason, so later results can be interpreted in context. If you change the offer, audience and destination at once, you may still improve the campaign, but you will have less evidence about which individual change mattered.
Turn the eight decisions into a one-page brief
Your brief should name the audience, outcome, primary channel, content themes, proof sources, destination pages, budget, owner and review date. Add the things you will not do: unverified claims, unsupported channels, excessive automation or content the team cannot maintain.
Check the brief with sales and delivery staff. They can identify promises the business cannot fulfil and questions the marketing team has missed. Then use the social media management workflow to translate the plan into weekly work, with the crisis guide available for escalation.
MattDarm's marketing strategy service can help define the plan, while social media management supports execution. Share your objectives and current activity if you want a scoped recommendation. The aim is a clearer route to useful enquiries, not a larger content quota for its own sake.
Frequently Asked Questions
What is the difference between a strategy and a calendar?
The strategy defines the audience, purpose, channels, proof and measurement. The calendar assigns specific posts, owners and dates to carry out that strategy. A busy calendar does not establish that the underlying choices are right.
How many channels should we start with?
Start with the smallest set you can support well, usually with a clear primary channel. Add another only when you have evidence of audience fit and capacity to produce suitable content and handle responses.
Do we need paid advertising?
Not automatically. Paid distribution can support a defined offer and audience, but needs its own budget, destination and measurement. It should not be used to compensate for an unclear proposition or a broken enquiry process.
What should our first report include?
Show the work delivered, production effort, relevant engagement, qualified enquiries and what happened next. Include measurement limitations and one or two decisions for the following period rather than a long list of vanity metrics.
When should we change the strategy?
Review it when the offer, audience or capacity changes, or when repeated evidence challenges the original assumptions. Individual post fluctuations usually call for investigation, not an immediate wholesale change.
