The Five Social Analytics That Matter (And Ten That Don't)

A post gets 20,000 views and your sales stay flat. Another reaches 600 people and brings in three enquiries, including one that becomes a £1,200 project. Which worked better? For a service business seeking clients, the second. Yet most social dashboards make the first look like the winner.

The problem is not a shortage of analytics. It is a shortage of priorities. Small businesses and creators need a short scorecard connecting attention to intent, customers and money. The five measures below do that. The other ten still have diagnostic uses, but none deserves to be your headline result on its own.

First, decide what a useful action looks like

Choose one primary outcome for each campaign before publishing. A local accountant might want qualified consultation requests. A maker might want first orders. A newsletter creator might want confirmed subscribers, then paid upgrades. Measuring all three as interchangeable “conversions” hides whether the content is doing its job.

Set up the minimum tracking: consistent UTM tags on campaign links, a recorded website conversion event, and a source field in your customer records. For enquiries arriving through messages, log the originating platform and post when known. Add “How did you first hear about us?” to your sales process; it catches journeys click tracking misses.

1. Qualified-action rate

This measures how often exposed people take an action that shows genuine intent. Count consultation requests that match your service, product enquiries with purchase intent, or confirmed email sign-ups. Do not count every emoji reply as a lead.

Formula: qualified actions ÷ people reached × 100.

If a campaign reaches 4,000 people and produces 12 qualified enquiries, its qualified-action rate is 0.3%. That is not automatically good or bad. Compare it with your own previous campaigns aimed at the same audience, promoting a similar offer.

Make the definition difficult to game

Write down what qualifies. For a wedding photographer, that might mean an enquiry containing a date, location and realistic budget. For a creator, it could be a confirmed subscription rather than a giveaway entry. Keep the definition stable so a rise reflects better intent, not looser counting.

2. Social-traffic conversion rate

Qualified-action rate starts with reach. This measure starts after the click: what percentage of visits from social produce your chosen website outcome? It helps separate a content problem from a landing-page problem.

Formula: completed target actions from social visits ÷ social sessions × 100.

Suppose a tutorial sends 240 sessions to a workshop page and generates 12 bookings. The conversion rate is 5%. A broad-interest video sends 900 sessions but generates nine bookings: 1%. The video brings more traffic, but the tutorial delivers more bookings from fewer visits.

Check the destination before blaming the content. A slow page, unexpected delivery charge or mismatched headline can waste strong intent. Segment by campaign and landing page; a platform-wide average can conceal the specific combination worth repeating.

3. Customer acquisition cost

Customer acquisition cost tells you what you spent to win each new customer. For a paid newsletter, count new paying subscribers. For an online shop, count first-time buyers, not every order placed by existing customers.

Formula: attributable acquisition spend ÷ new customers acquired.

Include advertising, freelance production, campaign tools and staff or owner time at a consistent internal hourly rate. If a month costs £300 in ads and £450 in production time, and brings 15 new customers, acquisition cost is £50. Calling it £20 because you counted only advertising makes the channel look artificially cheap.

Compare cost with contribution, not turnover

A £50 acquisition cost is uncomfortable if the first order leaves £25 after product and fulfilment costs. Repeat purchases might justify it, but use observed retention rather than optimistic lifetime-value estimates. For longer sales cycles, assess a cohort of leads once it has had time to become customers.

4. Revenue per 1,000 people reached

This connects distribution to commercial output. It is especially useful when a small, specialised audience buys more readily than a large, loosely interested one.

Formula: attributable revenue ÷ people reached × 1,000.

Imagine one campaign reaches 8,000 people and produces £640 in tracked sales. That is £80 per 1,000 reached. Another reaches 30,000 and produces £900: £30 per 1,000. The larger campaign brings more total revenue; the smaller one monetises its reach more effectively. You need both facts.

Use this within comparable campaigns, with the same attribution window. Do not add individual post reach figures and call the total unique campaign reach: the same person may see several posts. Where deduplicated reach is unavailable, label the denominator clearly and keep the method consistent.

A bigger audience is useful only when it improves the result you actually need.

5. Returning-customer or subscriber retention rate

Acquisition numbers can flatter content that attracts the wrong people. Retention tests whether social brings customers or subscribers who stay.

Formula: people in an acquisition cohort meeting your retention condition ÷ people originally in that cohort × 100.

Choose a condition matching the business. A paid membership might measure whether subscribers remain paid after 90 days. A coffee retailer might measure whether first-time buyers place another order within 60 days. A wedding business with little repeat demand could track referrals from past clients separately instead.

If 40 customers arrive through a social campaign and 10 reorder within 60 days, the cohort's repeat-purchase rate is 25%. Compare only cohorts that have completed the full window. Last week's customers have not had a fair chance to return.

The ten numbers that do not belong at the top

These metrics are not worthless. They answer narrower questions about delivery, creative or audience behaviour. Use them to investigate a business result, not replace one.

Build a scorecard you will actually use

Review monthly, with a brief weekly check for broken tracking or obvious campaign problems. Use one row per campaign and show the underlying counts beside each rate. One sale from ten visits is too fragile to declare a winning formula.

1. Record the offer, audience, spend and primary outcome.
2. Report the five measures where applicable; mark unavailable data honestly.
3. Use one consistent attribution model and conversion window.
4. Choose one change for the next cycle, such as a clearer offer or better landing page.

Keep tracked and self-reported sources distinguishable, and do not add platform-reported sales together without checking for duplication. Attribution is directional evidence, not proof that every credited sale happened because of social.

Conclusion: measure the next useful step

Keep the dashboard, but change the order you read it. Start with qualified action, conversion, acquisition cost, revenue efficiency and retention. Then use reach, views and engagement to explain what happened. That turns reporting into a decision about what to repeat, repair or stop.