By Chris Leo, Movou
I sell marketing, and my own organic social did almost nothing for two straight weeks. Twenty-six posts across Facebook, Instagram and LinkedIn, between the 1st and the 14th of September 2026. Zero clicks on two of the three platforms. I’m publishing the whole thing because every article you’ll find on whether organic social works for a service business was written by somebody selling organic social, and not one of them will show you their own numbers. So here are mine, zeros included.
One note before the numbers, because it matters for how you read them. Every figure below is a snapshot of that specific fortnight, read on the 14th of September. It is a record of what those two weeks produced, not a running scoreboard, and I have not re-pulled it since.
If you’re paying somebody 800 to 2,500 a month for this, you should know what the honest version of the output looks like.
Facebook: 8 posts, 869 impressions, zero clicks
Eight posts over the fortnight. 869 impressions. Zero clicks, zero reactions, zero shares, two comments.
Take a second with the shape of that. 869 people had our stuff put in front of them and not one of them tapped through. That’s not a disaster, it’s a diagnosis: we got distribution and we didn’t earn interest. Those are two different things and most reports only show you the first one.
Quick translation, because the platforms use these words loosely. Reach is how many separate people saw it. Impressions is how many times it got displayed, so one person scrolling past twice counts twice. Clicks is how many of them cared enough to do something. An agency report that leads with reach is choosing the biggest available number, and the biggest available number is the one that asks the least of your audience.
One more Facebook finding, and I want to be careful with it because it’s the kind of thing people turn into a rule far too fast. Over the previous two weeks on that same page, posts carrying a link in the body averaged 2.6 people reached. Posts without one averaged 266.4. That’s a hundredfold gap and it is very tempting to announce that Facebook buries links.
I’m not going to tell you that, because I haven’t proven it. There was no holdout, nothing else was controlled, and the sample is ten posts. What I have is a correlation big enough to be worth testing. So this week exactly one of our Facebook posts carries a link and the rest don’t, and I’ll report what happens after several cycles rather than after one.
Instagram: 14 posts, three comments, and a number I can’t see
Fourteen posts. Three comments. Zero likes, zero saves, zero shares. Account reach for the period: 53.
Fourteen posts is more than one a day and it moved essentially nothing.
Here’s the part that’s actually useful. Six of those fourteen were Reels, published between the 1st and the 8th, and they drew zero likes, zero comments, zero saves and zero shares between them. All three engagements the account got in two weeks landed on static posts.
Now the honest limit. I can’t currently pull per-post reach out of Instagram’s data connection, which means I cannot tell you whether those six Reels were seen and ignored or were never distributed in the first place. That distinction decides everything. If people saw them and scrolled, the hook is the problem. If nobody saw them, the account is the problem, and no amount of better hooks fixes it.
I don’t know which it is yet, so I’m not going to write you a confident paragraph about why video isn’t working for us. I’m cutting from four Reels a week down to two while I go find out, and that’s a spending decision, not a verdict on the format.
LinkedIn: 4 posts, 37 impressions, and the one that worked
Four posts. 37 impressions total, 33 of them unique people. Zero clicks, four likes, two comments, zero reshares.
Thirty-seven impressions in two weeks is a rounding error, and I’ll own that. But it’s the only platform of the three where the detail told me something I can act on.
The best post of the four ran on the 9th and recorded a 57.1% engagement rate on 7 impressions. Seven people saw it and four of them did something. Tiny sample, so treat the percentage as a hint rather than a result, but the content was a confession with a number in it, the same shape as this article.
The worst post ran on the 8th, carried a shortened LinkedIn link, reached 20 people (the widest distribution of the four) and recorded 0% engagement. Widest reach, no engagement at all.
I’ll also admit something unflattering here. Until last week our own reporting described LinkedIn as basically unmeasurable, because our field configuration was only pulling two data points. That was wrong. The connection exposes around 190 fields including per-post impressions, clicks and engagement rate. That was our setup, not a platform limit, and this is the first fortnight I’ve planned against real LinkedIn numbers.
The caveat that makes these numbers honest
Four days in this window (the 9th, 12th, 13th and 14th) reported zeros across every account in our reporting workspace, not just ours. That’s a reporting lag, not a fact about the world. So these totals may rise once those days fill in.
I’m telling you that up front because an article about honest numbers is the worst possible place to quietly present a partial total as a final one. Every figure above is what the data showed on the 14th of September, and I’d rather hand you the asterisk than have you find it later.
You’ll notice I also haven’t given you a single website traffic number. That’s deliberate. Our own analytics currently show the overwhelming majority of sessions as Direct, which is the bucket that means “we don’t know where this came from.” I don’t know the cause yet and I’m not going to guess at one in public. What I can say is that while that’s true, no session count on our site can honestly attribute anything to anything, so I’ve left them out entirely rather than dress them up.
So does organic social work for service businesses?
Here’s my read, and it’s a position rather than a finding.
Organic social is table stakes for most service businesses. Somebody who’s about to hire you will look you up, and a page that’s been dead for eight months costs you the job. A page that’s clearly alive doesn’t win you the job on its own, but it stops you losing it. That’s real value and it’s worth about one post a week, not fourteen.
What it is not, at the volume most people are sold, is a lead channel. Twenty-six posts bought us two comments, three comments and four likes across three platforms. If somebody is charging you two grand a month and showing you reach, ask them for clicks. Then ask them for booked jobs. The number gets smaller each time you ask, and the one at the end is the only one that pays anybody.
Key takeaways
- Reach measures distribution. Clicks measure interest. Only one of them can turn into a customer, and reach is the number that gets put in front of you.
- Pick the metric before you publish, not after you see the result. A number chosen afterwards is a story about the result, not a measurement of it.
- Don’t let a big correlation become a rule. Our link posts reached a hundredth of our link-free posts, and that’s still only a reason to run a test.
- Keep the channel alive because buyers check it. Don’t fund it like a lead source until it behaves like one.
What I’m changing
Reels drop from four a week to two while I find out whether the six that flopped were seen at all. One Facebook post this week carries a link and the rest don’t, so the correlation gets a test instead of a headline. LinkedIn gets planned against real numbers for the first time, including one post in a format we’ve never run on that account.
That’s it. No relaunch, no new strategy deck. Four small changes with a number attached to each one and a date to review them.
If you want the same treatment on your own numbers, honest version included, here’s how we handle social, or book a call and we’ll pull your actual figures together.
What number does your marketing report lead with?
Further reading: Sprout Social maintains an ongoing roundup of social media ROI benchmarks if you want the industry-wide picture next to this one, and Google’s Search Console documentation explains how impressions and position are counted on the search side.