I Tracked One Hashtag Every Day for a Week — Here's What the Data Showed

A single snapshot of a hashtag is almost useless. You see a pile of posts, some with big numbers, and you can't tell what's rising, what's stale, or who actually drives the tag. So I ran a small experiment: pick one hashtag, pull it at the same time every day for seven days, store everything, and see what a week of movement reveals that one snapshot can't.

I'll walk through the setup, the patterns that showed up, and the honest caveats (the numbers below are from one representative run, yours will differ, that's the point of running your own). This is less "here's a definitive study" and more "here's a cheap experiment that tells you more than any trending page."

The setup

The whole thing is a scheduled call plus a stored row per day. I used the Instagram and TikTok hashtag-search endpoints from SociaVault, one credit per pull, so a week of daily pulls across both platforms costs 14 credits. New accounts get 50 free, which covers this with room to spare.

const API_KEY = process.env.SOCIAVAULT_API_KEY;
const BASE = "https://api.sociavault.com/v1";

async function pullInstagram(hashtag) {
  const res = await fetch(
    `${BASE}/scrape/instagram/search/hashtag?hashtag=${encodeURIComponent(hashtag)}&date_posted=last-day`,
    { headers: { "x-api-key": API_KEY } },
  );
  return Object.values((await res.json()).data?.posts ?? {});
}

function daySummary(posts, day) {
  const likes = posts.map((p) => p.like_count ?? 0);
  const total = likes.reduce((a, b) => a + b, 0);
  const topByLikes = [...posts].sort((a, b) => (b.like_count ?? 0) - (a.like_count ?? 0))[0];
  return {
    day,
    postCount: posts.length,
    totalLikes: total,
    medianLikes: median(likes),
    topCreator: topByLikes?.owner?.username ?? null,
    topLikes: topByLikes?.like_count ?? 0,
  };
}

function median(nums) {
  if (!nums.length) return 0;
  const s = [...nums].sort((a, b) => a - b);
  return s[Math.floor(s.length / 2)];
}

Filtering to last-day keeps each pull focused on fresh posts, so day-over-day counts mean something. I ran daySummary each morning and appended the row to a small table. That's the entire apparatus. The hashtag search walkthrough covers the fields in more detail.

What a week actually showed

Here's the shape of a representative run for a mid-size niche tag (again: illustrative, run it on your own tag for real numbers):

• Volume wasn't flat, it had a rhythm. Post counts weren't random. There was a clear weekday lull and a weekend bump, the kind of cadence you'd never see from a single snapshot. Knowing your niche posts more on Saturdays is a genuinely useful scheduling insight.
• The median told a different story than the total. Total likes spiked on days when one post went big, but the median barely moved. That gap, high total, stable median, is the signature of one viral outlier riding an otherwise ordinary day. If you only tracked totals, you'd misread a fluke as a trend.
• A handful of creators kept reappearing. The same few usernames showed up in the top slot across multiple days. Seven daily pulls surfaced the niche's actual regulars far better than one big pull, because consistency is only visible over time.
• New entrants stood out. On day five a creator I hadn't seen all week landed the top post. A brand-new account punching above its follower weight is exactly the "catch it early" signal people want from trending, and it only pops out against a baseline.

None of that is visible in a one-time scrape. The value came entirely from the repetition.

The three metrics worth storing

If you run this, don't just keep the raw posts. Compute and store these each day, they're what make the week readable:

1. Post count (fresh only). Volume trend for the tag. Rising volume over several days is a heating niche.
2. Median engagement, not just total. The median resists outlier distortion. Watching it climb is a better "the whole tag is getting more traction" signal than totals.
3. Top creators, deduplicated across days. A frequency count of who lands top posts. This is your niche's regulars list, built automatically.

function regulars(weekRows) {
  const counts = {};
  for (const row of weekRows) {
    if (row.topCreator) counts[row.topCreator] = (counts[row.topCreator] ?? 0) + 1;
  }
  return Object.entries(counts).sort((a, b) => b[1] - a[1]);
}

After a week, regulars hands you a ranked list of who consistently wins the tag, which is a creator shortlist you didn't have to build by hand.

Run it on TikTok too, and compare

The same experiment on TikTok's hashtag search (pass the tag without the #) is worth running in parallel:

async function pullTikTok(hashtag, region = "US") {
  const res = await fetch(
    `${BASE}/scrape/tiktok/search/hashtag?hashtag=${encodeURIComponent(hashtag)}&region=${region}`,
    { headers: { "x-api-key": API_KEY } },
  );
  return (await res.json()).data;
}

Reading fields defensively matters more here, TikTok's payload shape differs from Instagram's, so log one response and confirm before trusting a field. Running both lets you see where a tag is more active, which platform your niche's momentum actually lives on. That cross-platform read is often the most actionable output.

The honest caveats

• It's a sample, not a census. Each pull returns a slice of public posts, not every post under the tag. Trends across days are meaningful; exact counts aren't gospel.
• The numbers above are illustrative. I'm showing the shapes a week reveals, not publishing a fixed dataset. Your tag, region, and week will produce different figures, run it yourself for numbers you can trust.
• Indexed public data has lag. The Instagram side leans on Google-indexed public posts, so the very newest content may not appear instantly. Fine for daily trends, not for second-by-second alerts.
• One week is a starting point. Seven days shows rhythm; a month shows seasonality. The method scales, just keep appending rows.

I'd rather you treat this as a repeatable method than a headline stat. The point isn't my numbers, it's that a trivially cheap daily pull turns a hashtag from a static pile into a moving picture.

Who should run this

• Content teams figuring out when their niche actually posts and what formats are climbing.
• Influencer scouts who want a niche's regulars and new entrants surfaced automatically instead of scrolling.
• Trend researchers who lost easy access to platform trending pages and want a self-built tracker for the tags they care about. It complements finding trending content on any platform.

Frequently Asked Questions

What can tracking a hashtag over a week tell me?

Movement that a single snapshot hides: posting rhythm across days, whether median engagement is genuinely rising or just spiking on outliers, which creators consistently win the tag, and when a brand-new entrant breaks through. The repetition is where the insight lives.

How much does a week-long hashtag study cost?

One credit per pull. A daily pull on one platform is 7 credits for the week; running Instagram and TikTok in parallel is 14. New accounts start with 50 free credits, so a first study costs nothing.

Why track median engagement instead of total likes?

Because totals get distorted by a single viral post. The median stays stable unless the whole tag is genuinely gaining traction, which makes it a more honest signal of a rising niche versus a one-off fluke.

Is this real-time?

No. The Instagram side relies on Google-indexed public posts, so there's a lag before brand-new content appears. It's built for daily trend tracking, not instant alerts. TikTok search is closer to live but still returns a sample.

Can I track more than one hashtag?

Yes, and you should. Extend the watch list to the 10 to 30 tags that define your niche and store a row per tag per day. That turns the experiment into an ongoing niche dashboard.

Do the numbers in this post reflect a specific hashtag?

They're illustrative shapes from a representative run, not a fixed published dataset. The method is the deliverable, run it on your own tag and region for numbers you can actually cite.

Wrapping up

A hashtag snapshot is a photo; a week of snapshots is a film. For the price of a few credits a day, you can watch a niche's posting rhythm, tell real momentum from outlier spikes, and get a regulars-and-newcomers list built for you. Skip my example numbers, the real payoff is running it on the tag you actually care about and watching it move.

Want to run your own week-long hashtag study? Start free with SociaVault, 50 credits, no card, and log your first day in a couple of minutes.

---

Related Articles

• How to Search Instagram by Hashtag Without Logging In
• Why TikTok Trending Moved Behind a Login (and How to Still Track What's Rising)
• How to Find Trending Content on Any Social Platform
• TikTok Keyword Search for Content Research