Constant Contact’s Q2 2026 “Small Business Now” report puts consumer discovery of new small businesses at 49% on social media against 40% through search engines, which the company describes as social having overtaken search.
That figure is going to get quoted at you this year. What sits underneath it is less flattering, and worth reading before you change anything.
The discovery flip
Constant Contact surveyed more than 5,000 small business owners and consumers across the US, UK, Canada, Australia and New Zealand for its June 2026 report. 49% said they discover new small businesses on social media. 40% said search engines.
Two caveats. It is a five-country sample, not a US reading, and Constant Contact sells marketing software, which does not make the number wrong but does mean the company has a stake in the answer.
A US study measuring something adjacent lands elsewhere. BrightLocal asked a representative panel of 1,000 US adults in April 2025 where they go first for a local search. 45% default to Google, 20% to a maps app, 14% to social platforms.
The two are not in conflict, because they asked different questions. BrightLocal asked where you start when you already know you need a plumber. Constant Contact asked where you come across a business you had never heard of. A customer can meet you on Instagram Tuesday and still type your name into Google Thursday before she calls.
So the finding is narrower than the headline: in a five-country vendor survey, social is now the more common place for first contact with an unfamiliar small business. It does not establish that search matters less once someone is ready to hire.
The engagement numbers nobody wants to hear
Quid, formerly Rival IQ, published its 2026 Social Media Industry Benchmark Report in March, drawing 150 companies at random from each of 18 industries out of a database of more than 200,000. It reports medians rather than averages, which stops one runaway account from lifting a whole industry. Quid also discloses using its own AI agents to analyze and synthesize the data. Engagement rate is total interactions divided by follower count.
- TikTok 2.01%
- YouTube 0.21%
- Instagram 0.30%
- Facebook 0.15%
- X 0.03%
These are percentages of your own followers, so the arithmetic is worth doing out loud.
A Facebook page with 1,000 followers, at the 0.15% median, gets about 1.5 interactions on a typical post. One like and a half. Post three times a week and the whole week produces four or five. The same 1,000 followers on Instagram, at 0.30%, gives you 3 interactions per post. On X, 0.03% works out to 0.3 per post, or one interaction roughly every third time you post.
If you have been posting all year and getting two or three likes on a page with a few hundred followers, you are not doing it wrong. You are sitting on the median. Quid also found Instagram’s median falling from 0.36% to 0.30% year over year, a drop of about 17%, so the bar moved down while you were trying harder.
Two credible studies disagree, and what to do about that
Socialinsider’s Social Media Benchmarks For 2026, published in January with a quarterly update in July, analyzed 70 million posts from international brands across TikTok, Instagram, Facebook and X between January 2024 and December 2025. Its 2025 figures: TikTok 3.73%, Instagram 0.48%, Facebook 0.15%, X 0.12%. The mid-2026 update: TikTok 2.70%, Instagram 0.45%, Facebook 0.13%, X 0.10%.
Put Instagram side by side. Socialinsider says 0.48%. Quid says 0.30%. On a 1,000-follower account, 4.8 interactions per post against 3.0. Both organizations publish their methods.
Neither is cheating. The first reason is sample composition. Quid drew randomly across 18 industries; Socialinsider worked from posts by international brands. Higher education does not behave like fast food, and a random draw across 18 industries is a different population from a body of international brand accounts.
The second reason is the definition of an interaction. Both divide by follower count, but the analyst decides what goes in the numerator. Saves, shares, video views, comment replies: include or exclude any of them and the rate shifts. A small difference in what lands inside the word “interactions” produces the gap you are looking at.
Which points to a rule. Do not measure yourself against an industry median drawn from a sample you are not part of. Measure yourself against your own trend line.
Add up interactions across your last 90 days of posts, divide by follower count, then divide by the number of posts. Run it the same way every quarter. Whether that rate beats a published median tells you little, because you were never in that sample. Whether it beats your own last quarter tells you whether the work is doing anything.
The two studies do agree on direction. Socialinsider found average comments per post down 24% on TikTok and between 16% and 20% on Instagram. Quid found Instagram’s median down about 17%. Different samples, different definitions, same slope.
Posting frequency hit a six-year low
Quid’s report also puts median posting frequency at 3.7 posts per week on Instagram and 3.0 per week on Facebook, both the lowest in its six-year series. TikTok held at 2.0 videos per week, X at 2.24.
If you have been treating your own shortfall as a discipline problem, the bar has been dropping for six years, and companies with marketing staff are the ones setting it.
The line most likely to change your plan is a ratio: in Quid’s numbers, TikTok generated the highest engagement of any platform on roughly half Instagram’s post volume, 2.0 videos a week against 3.7 posts. Volume is not the lever most owners assume.
The platform nobody expected to come back
Metricool’s 2026 Social Media Study, published in January, compared 39,762,999 posts across 1,059,949 accounts on ten platforms, measuring organic performance year over year. Facebook reach rose 51%, impressions rose 57%, interactions rose 56%.
Instagram moved the other way in the same dataset. Reels reach fell 35% and post reach fell 31%.
This is one company’s account base, and its users skew toward businesses already running a scheduling tool. Still, a 51% reach increase across nearly 40 million posts cuts against the assumption that Facebook is where posts go to be ignored.
Notice what did not move. Facebook engagement per follower sat at the bottom of both benchmark studies: 0.15% from Quid, 0.15% falling to 0.13% from Socialinsider. Reach went up. The share of followers who press anything did not.
Where the hour goes
LocaliQ’s 2026 small business marketing report, from a survey of more than 300 owners with 87% in the US or Canada, found 60% spend 1 to 10 hours a week on all marketing, rising to 72% at firms with 10 or fewer employees. Social gets a slice of that. Call it an hour.
Twenty minutes, one platform, three posts. Three a week puts you level with Quid’s Facebook median of 3.0 and near its Instagram median of 3.7, the numbers companies with marketing departments are hitting. Pick the platform your customers already use. If Metricool’s reach data describes your audience, Facebook is more defensible this year than two years ago. Write all three in one sitting and schedule them.
Twenty minutes, one video. Quid’s finding that TikTok produced the highest engagement on about half Instagram’s post volume is the strongest case in this data for making fewer, heavier things. One video a week, shot on your phone, is a real commitment inside one hour, and the slot most likely to return something. The same clip goes to Reels and TikTok.
Fifteen minutes, replies. Comments per post are falling on both TikTok and Instagram in Socialinsider’s numbers, which makes the ones you get worth more. Answer all of them, including the messages in the folder you rarely open.
Five minutes, the number. Write down interactions, followers and post count. Once a quarter, work out your rate and set it beside last quarter’s.
Not on the list: a fifth platform, or daily posting on the one you have. The posting frequency data supports neither for a business with one hour.
Constant Contact also found that 47% of small business owners handle all their social media on their own, and 73% now describe themselves as content creators to some degree. Plenty are doing careful work. Whether it belongs in the owner’s hour is a different question from whether the owner can do it.
You can run your own social media. If that hour is one you look forward to, keep it. If it is the hour you spend at 10pm feeling behind, price it against everything else it could have been. When social is part of the digital marketing work we take on, the first month is unremarkable: one platform, a cadence the business can hold, expectations set against your own trend rather than a stranger’s median. That is the difference between still posting in month nine and quietly stopping in month three.
One step this week. Run that calculation once, on your strongest platform, over your last ten posts, and write the number down with today’s date. Then set a reminder for ninety days out to run it again the same way. That is your baseline, and without it nobody, including us, can tell you whether anything you do next is working.