Social Media Analytics for Small Business: What to Watch

Social media analytics for a small business should answer three questions and then get out of the way: did my posts actually go out, is anything changing, and is there one thing I should do differently next week. You do not need a dashboard with forty tiles. You need a handful of figures you can trust, recorded over time, that you look at once a week for ten minutes.

This guide is for the owner who runs the café, the salon, the practice or the trade, and does the social media too. It covers which numbers matter, which do not, what the words mean, and how to keep the whole thing to a weekly check.

Start with the question analytics is for

Analytics for a small business is not about proving anything to anyone. Nobody is asking for a report. It is about making next week's posting slightly better than this week's, and noticing when something has broken. Any figure that does not serve one of those two purposes can be ignored, whatever the dashboard thinks.

The three figures that matter

1. Did it publish?

The first number is the least glamorous and the most important: how many of the posts you meant to send actually went out, on each platform. A connection that quietly expired two weeks ago is the most common reason a small business "went quiet", and the owner is usually the last to know. Look for failed and unconfirmed posts before you look at anything else. Make sure your tool shows them rather than counting the attempt; why failed social posts should not appear as published explains what to check.

2. Is it changing?

The second number is a comparison: this period against the one before it. Posts published, and whatever engagement each platform gives you, for the last 7 or 30 days against the previous 7 or 30. A total on its own tells you nothing. A change tells you whether what you did last month worked. That needs a history, recorded over time, which is the one thing the platforms themselves will not keep for you; see how to build a history of social media performance.

3. Which post did the work?

The third is the post-level view: of the posts that went out, which ones people responded to. Not to chase a formula, but to notice patterns. The photo of the team outperformed the price list. The Tuesday post did better than the Saturday one. One observation like that a week is plenty.

The figures to ignore

• Cross-platform totals. Adding one platform's views to another's impressions gives a number with no meaning. Reach vs impressions vs views vs engagement explains why the words do not line up.
• Minute-by-minute numbers. Watching a figure move as it happens is a hobby, not analysis. Weekly is the right cadence for a business that has other things to do.
• Vanity trends drawn from one point. A chart rising from a single reading is decoration. Two readings make a line; one does not.
• Zeros you did not measure. A figure the tool could not read should say so. A zero that means "unknown" will send you looking for a problem that is not there.

What the words mean, briefly

You do not need to memorise the vocabulary, but four words come up constantly. Reach is roughly how many different people saw a post. Impressions is roughly how many times it was shown, including to the same person twice. Views is what video-led platforms count, with each platform deciding how long counts as a view. Engagement is people doing something: liking, commenting, sharing, saving. Each platform defines each of these its own way, which is why they are compared within a platform and not across.

How often to look, and for how long

Once a week, ten minutes, same day each week. Longer than that and you start reading meaning into noise. Less often and a broken connection runs for a fortnight. The routine:

1. Anything failed or unconfirmed? Fix it first.
2. Last 7 days against the previous 7: posts and deliveries, by platform.
3. Which one or two posts did best? Why, in one sentence?
4. One thing to do differently next week. Write it down.

How WyreIt keeps the numbers honest

WyreIt Analytics is built so that a weekly check is enough, because the figures do not need second-guessing.

• Published means confirmed. A post counts as published only when the platform has handed back its own record. Failed destinations are shown as failed; unconfirmed ones as unconfirmed. If one platform published and another failed, WyreIt records exactly that. What happens when one platform publishes and another fails shows what that looks like.
• 7, 30 and 90 day views, each against the period before. So the change is on the screen, not in your head.
• Per platform and per account. Two accounts are two histories. Nothing is added across platforms into a number with no unit.
• Measured zero is 0; unread is unavailable. With the reason where there is one, and "unsupported" where a platform simply does not offer the figure.
• A history that accrues in the background. You do not have to open Analytics for the readings to be taken. Where a platform supports an account-level figure such as followers, it is recorded daily and becomes a trend once there are enough readings to draw one.
• Freshness on every figure. Each number says when it was taken.

Analytics is part of every WyreIt plan and works during the 14-day free trial. There is no analytics add-on and nothing to switch on. The page for WyreIt Analytics has the full account, including what it deliberately does not do.

What analytics will not do for you

It will not tell you what to post. WyreIt Analytics shows figures; it does not generate recommendations from them. It is not real-time, and it cannot show history from before the account was connected. And it cannot make a post good. What it can do is make sure that when you look, once a week, the numbers you see describe what actually happened, so the one decision you make is based on something real. How to track social media performance across multiple platforms is the next step once the weekly habit is in place.

Common questions

Do I need a separate analytics tool as a small business?

Usually not. The platforms show you today's figures for free, and a scheduling tool with honest analytics records them over time. What you need is publication truth, a comparison with the previous period, and a history. If your current tool gives you those, you have enough.

How many followers should I have?

The useful question is whether the number is moving, in the right direction, for your account, over time. A café with a few hundred local followers who come in is doing better than one with thousands who do not. Compare your account with its own past, not with anyone else's.

Is analytics included in the WyreIt trial?

Yes. Analytics is on every plan and works during the 14-day free trial, with no card needed. The trial does not include WyreIt AI, which is a separate matter from analytics.

What if I run two locations on the same platform?

Connect both accounts and read them separately. WyreIt keeps each account's figures and history apart, so the second branch is never blended into the first.

If WyreIt says it published, it has evidence it published. Three questions, once a week, from figures you do not have to second-guess. Analytics is included on every plan and in the 14-day free trial, with no card needed. See how WyreIt Analytics works or see pricing.