Skip to content

Troubleshooting

Why Is My Upload Speed Slow?

8 min readPublished

Why does sending things take so long?

Sending things out of your home takes far longer than getting things in.

Does this sound familiar?

  • Photos and files take ages to send.
  • Others say you freeze on calls, but they look fine to you.
  • Backups run all night and still are not done.

This page explains why sending is slower, and what you can change.

What happens if I ignore this?

If you do nothing, backups and shared files keep taking most of the night.

The short answer

Most home plans are built to receive far more than they send, so sending is slower on purpose.

Try this first

  1. Check your plan and see what sending speed you were sold.
  2. Send large files overnight, when nobody else needs the connection.
  3. Use a cable for calls and big uploads if you can.
Run a test now

Still want to know why this happens? Keep reading.

Short answer

Upload speed is usually much lower than download speed on home connections, because most plans are built asymmetrically around typical usage. A genuinely disappointing upload result, though, can also come from a saturated upstream link, competing uploads from other devices, a weak Wi-Fi connection, a VPN, or temporary provider congestion. A single test cannot tell these apart, and it is never enough evidence on its own to conclude your provider is at fault.

Key takeaways

  • Most home connections are asymmetric by design, with far less upload capacity than download capacity.
  • Video calls, cloud backup and sharing photos or videos are the everyday activities most affected by a slow upload.
  • Competing uploads from other devices, Wi-Fi limitations and a VPN can each explain a poor result independently of your provider.
  • Upload-loaded latency — delay that appears specifically while uploading — is a separate, related measurement worth checking alongside raw upload speed.
  • One upload test is not enough evidence to conclude anything about your provider; a consistent pattern across several tests is needed.

Why upload and download are treated so differently

Download and upload are two separate measurements because they usually travel over separate capacity, and because most home internet usage is heavily download-weighted. Streaming video, loading web pages and downloading files all pull far more data than the average household sends back out, so providers have historically built plans that reflect that pattern.

Asymmetric connection
A connection where download capacity and upload capacity are set to different levels, with download almost always the larger of the two. This is standard on cable and many fibre and DSL plans, and it is a deliberate design choice rather than a fault.

This means a large gap between your download and upload numbers is often completely normal and expected, not evidence that something is wrong. The relevant question is not 'why is my upload lower than my download', which usually has a simple structural answer, but 'is my upload speed lower than what my plan actually promises', which is a very different question with a very different set of possible causes.

Where a slow upload actually shows up

Upload speed rarely matters for browsing or streaming, which is why it can go unnoticed for a long time. It becomes noticeable the moment you do something that sends a meaningful amount of data outward rather than pulling it in.

Video calls are the most common everyday case: your outgoing video and audio stream needs a steady amount of upload capacity for the whole duration of the call, and if that capacity is not reliably available, the other participants see your video freeze, drop in quality, or your audio break up — even if your download speed is excellent.

Cloud backup services are another frequent cause of confusion, because they can run quietly in the background and consume a large share of available upload capacity without any obvious indication on your screen. Sharing photos or videos from a phone, syncing a large folder, or backing up a laptop overnight can all saturate a modest upstream link for extended periods.

Upstream saturation and competing upload traffic

Because upload capacity is usually the smaller of the two directions, it takes far less activity to fill it completely. A single large backup or a video call from one device can be enough to use most or all of the available upstream capacity on some plans, leaving very little headroom for anything else happening at the same time.

If more than one device tries to upload at once — one person on a video call while another device is backing up photos, for example — they compete directly for the same limited capacity, and both suffer as a result. This looks, from a test result alone, exactly like a genuinely slow connection, even though the underlying cause is simply too much demand for too little upstream capacity at that moment.

Pausing other uploads and retesting is one of the fastest ways to check whether this is happening, because if the result improves substantially, competing traffic on your own network was a meaningful part of the picture.

Wi-Fi, VPNs, and provider-side limitations

Wi-Fi can constrain upload just as it constrains download. A weak signal, a device far from the router, or a busy wireless channel with many nearby networks can all reduce the effective upload speed available to a device, independent of what your plan actually provides. Testing over a wired Ethernet connection removes this variable entirely.

A VPN adds encryption and typically routes traffic through a specific exit server, and this overhead applies to uploads as well as downloads. If that server is distant, busy, or itself capacity-constrained, your effective upload speed through the VPN can be noticeably lower than your connection's real capacity. Testing with the VPN temporarily disabled is a quick way to check its effect.

Finally, upstream capacity can be temporarily limited by your provider, particularly at peak usage times when many households in a shared area are active at once. This is a plausible explanation for a pattern that is worse in the evening and better early in the morning, but it should be treated as one hypothesis to check, not an assumption to start from — the other, more local causes above are all easier and cheaper to test first.

Upload speed and upload-loaded latency are related but separate

Raw upload speed tells you how much data your connection can send per second at full capacity. It does not tell you what happens to responsiveness while that capacity is being used. That is a separate measurement: upload-loaded latency, which describes how much delay appears specifically during an upload compared with when the connection is idle.

A connection can have a perfectly adequate raw upload speed and still produce a bad video call experience, if uploading causes a large increase in response time. In that situation, the fix is less about capacity and more about how equipment on the path manages the queue of data waiting to go out — smart queue management on a capable router is the most directly relevant setting to check.

If your video calls degrade specifically while a backup or large upload is running, but seem fine otherwise, upload-loaded latency is a more likely explanation than raw upload speed being insufficient.

What slower upload speeds mean in practical terms

It can be easier to judge an upload speed by thinking in terms of how long realistic files actually take to send, rather than by the megabit-per-second number alone. The table below gives rough estimates for a few common upstream speeds, assuming the full speed is available and nothing else is competing for it.

Estimated upload time for common file sizes at different upstream speeds
FileApprox. size5 Mbps upload20 Mbps upload100 Mbps upload
A few phone photos20 MB~32 seconds~8 seconds~2 seconds
Short phone video clip200 MB~5.3 minutes~1.3 minutes~16 seconds
An hour of HD video footage3 GB~80 minutes~20 minutes~4 minutes
A full phone backup20 GB~9 hours~2.2 hours~27 minutes

These figures assume the full measured upstream speed is available for the entire transfer, which in practice is rarely the case once other devices, background activity or Wi-Fi limitations are taken into account. They are meant as a rough, practical reference for what a given upload speed actually means for the kind of files people send day to day, not a guarantee of any particular transfer time.

How this shows up in your Norynt result

Upload speed
The raw capacity your connection can send data at, tested independently of download speed.
Upload-loaded latency
How much response time increases specifically while an upload is running, which affects video calls even when raw upload speed looks adequate.
Throughput consistency
Whether upload speed holds steady during a transfer or varies significantly, which affects how reliable large uploads and calls feel in practice.
Idle latency
The baseline responsiveness of a quiet connection, useful for confirming that a slow upload is not accompanied by an unrelated general latency problem.
Run a Norynt test

What this means in everyday use

Video calls
A constrained upload directly affects how your own video and audio appear to other participants, regardless of how fast your download speed is.
Cloud backup
Backups can take drastically longer than expected on a modest upstream link, and can quietly consume capacity needed by other activities at the same time.
Sharing large files
Sending large photos, videos or documents to colleagues or family can be slow even on a fast download plan if upstream capacity is limited.
Remote work
Uploading files to shared drives or presenting your screen depends on upload capacity, and can lag during other simultaneous uploads on the network.

Check it for yourself

Change one condition at a time and measure again. A result that improves supports an explanation; it does not prove it.

  1. Pause heavy traffic and retest

    Pause other uploads and background syncs on every device, then retest, to check whether competing traffic was limiting your result.

  2. Test over Ethernet

    Test with a cable directly into the router to check whether Wi-Fi is constraining your upload speed.

  3. Disable the VPN temporarily

    Disable any active VPN temporarily and retest to check whether it is adding overhead to your uploads.

  4. Test at another time

    Test again at a different time of day to check for a pattern consistent with peak-time provider congestion.

What should I do?

  1. Pause backups and background syncs before testing

    Free and immediate — removes the most common source of a misleadingly low upload result.

  2. Test over Ethernet if possible

    Rules Wi-Fi in or out as a factor in a couple of minutes.

  3. Compare your result with your plan's advertised upload speed

    A gap only matters relative to what your plan promises, not relative to your download speed.

Things people get wrong

Comparing upload speed directly with download speed and assuming something is broken.
Most home plans are asymmetric by design, so a large gap between the two numbers is expected and not, by itself, a sign of a problem.
Concluding the provider is at fault after one slow upload test.
Competing traffic, Wi-Fi, a VPN and background backups are all more common and more local causes, and are worth ruling out first.
Assuming a background backup finished or isn't running.
Cloud backup and sync tools can run quietly for hours and consume most of the available upload capacity without any obvious on-screen indication.

Common questions

Why is my upload speed so much lower than my download speed?

Most home internet plans are built with far more download capacity than upload capacity, reflecting typical usage patterns like streaming and browsing. A large gap between the two is normal on most connections and is not, by itself, evidence of a fault.

Can a background backup really slow down a video call?

Yes. If a backup is using most of the available upstream capacity, there is little left for your call's outgoing audio and video, which can cause freezing, dropped quality or garbled audio for the people you're talking to.

Does a VPN affect upload speed as much as download speed?

It can affect both, since the same encryption and routing overhead applies in each direction. The exact impact depends on the VPN service and the distance to its exit server.

Is a slow upload always the provider's fault?

No. Competing devices, Wi-Fi limitations, a VPN and background uploads are all common local causes that should be checked first. A provider-side limitation is one possible explanation among several, not the default assumption.

What is upload-loaded latency, and why does it matter here?

It measures how much your connection's response time increases specifically while an upload is running. A connection can have adequate raw upload speed but still cause laggy calls if this figure is high, because the two measure different things.

Measure it yourself

A Norynt test measures download, upload, idle latency, jitter and latency under load, then explains what the numbers mean for the things you actually do.

Run a test

Norynt writes its own explanations. External documentation is listed only so you can verify what you read here.

Written by Norynt Editorial. Checked by Norynt Measurement Team.

Published

We correct pages when the facts change. Tell us if something here looks wrong.

Technical references

The information in this article is based on publicly available technical documentation and networking research. You do not need to read any of it to understand the article.

Measurement Lab(1)

The open measurement platform Norynt runs its speed tests on.

RFCs(1)

The formal internet standards behind these terms.

Browser documentation(1)

What a browser can and cannot measure in a web page.

Standards and industry documentation(1)

Industry guidance from equipment makers and service providers.