Slowpoke vs Tideways
Tideways is the pragmatic PHP choice: monitoring and a callgraph profiler in one hosted product, with request-based pricing and a European company behind it. Slowpoke overlaps with the monitoring half, costs nothing, stays on your machines, and does not stop at PHP.
The short answer
Which one you should pick
If your estate is PHP and you want profiling and monitoring from one vendor, Tideways is a sensible purchase. The trade is the meter, the hosting, and everything that is not PHP.
Choose Tideways if
You are a PHP shop and want profiling included
- You want a callgraph profiler and production monitoring from the same product, without wiring two tools together.
- You are on Symfony, Laravel, Magento, Shopware or TYPO3 and want instrumentation that already knows those frameworks.
- You want exception tracking and alerting alongside performance.
- You would rather buy it hosted, and a European vendor matters to you.
Choose Slowpoke if
You want no meter, and more than PHP
- Some of your estate is not PHP — a Django service, a Go binary, a Next.js front end.
- The database server, Redis, the PHP-FPM pools and the queue machines are part of your problem.
- You do not want a monthly request limit to be a thing you think about.
- Your data has to stay on your machines.
- You want the expensive findings prioritised in business terms and turned into work that closes.
Side by side
Slowpoke and Tideways, line by line
Written to be fair, including where Tideways is the better tool. If a row here is wrong or out of date, that is a bug: open an issue and it gets corrected.
| Aspect | Slowpoke | Tideways |
|---|---|---|
| Price model | €0, self-hosted, no meter | Monthly PHP request volume |
| Entry price | €0 | Team from about €29 a month at 250,000 requests; Business from about €89; Flex from about €69 with unlimited applications (list, 2026) |
| Cost of more traffic | €0 | A higher request tier |
| Where your data lives | Your machines, always | Tideways' cloud |
| Languages | Any, through logs, DB statistics and OTLP | PHP |
| Callgraph profiling | No | Yes, and it is a main reason to buy it |
| Code origin of a query | File and line, 2.0 µs per query, no extension | Yes, via its PHP extension |
| Needs a PHP extension | No | Yes |
| Query plans with a fix | Real EXPLAIN, twelve named diagnoses | Shows the statement |
| Priced by business cost | Seconds of waiting per day | Slowest and most time-consuming first |
| The machine and the database server | CPU, iowait, swap, OOM, disk, Redis, PHP-FPM pools | Application, plus some server metrics |
| Maps what you inherited | One command reads each machine | — |
| Work that verifies itself | Missions closed by measurement | — |
| Exception tracking | Not its job | Yes |
No extension, and a number you can reproduce
Tideways does its deepest work through a PHP extension, which is how it gets a callgraph. That is the right engineering for a profiler, and it is also a thing you have to install on every machine, keep in step with your PHP version, and explain to whoever signs off on production changes.
Slowpoke's Laravel package is ordinary PHP: no extension, no OpenTelemetry SDK, nothing to compile. It costs 2.0 µs per query on PHP 7.4 and 1.9 µs on PHP 8.3 — 0.10 ms on a request with fifty queries, roughly one ten-thousandth of a request that spends 800 ms in your code and your database. Run ./bin/bench and you get those numbers on your own machine. It sends after the response with a hard 0.1 second budget: a missing agent costs you one trace, never a request.
It also does not need to be everywhere. Access logs, the php-fpm slow log and database statistics already give you endpoints, queries and costs with no application change at all; the package is the upgrade you add when you want the file and the line.
The part of the estate that is not PHP
Almost nobody is only PHP any more. There is a Python service doing reports, a Node process somewhere, a Go binary somebody wrote in a weekend, a WordPress install marketing owns, and the database server everything shares. A PHP-only tool draws a boundary around the part of the problem that happens to be in PHP, and the slowest thing is not always inside it.
Slowpoke puts all of it on one scale. Any backend works through access logs, database statistics and OpenTelemetry, and the agent recognises Laravel, Symfony, WordPress, Django, FastAPI, Flask, Next.js, NestJS, Express, Koa, Go and Rails when it reads a machine — then tells you which of them it cannot fully read yet, and the lines that would fix that.
Monitoring priced the way a business thinks
Tideways ranks by the usual things: slowest transactions, most time consumed. Useful, and still engineer-facing. Slowpoke prices in seconds of user waiting per day and carries that number all the way up: every item gets an effort estimate and a return per hour of work, and the debt report adds a weekly repayment budget and delivery health — deploy frequency, change failure rate, time to restore.
That is the page you send to the person who decides whether next sprint is features or performance, and it is written before you ask.
Can you run both?
Yes. If you already pay for Tideways, keep the profiler — Slowpoke has nothing like it — and let Slowpoke cover the non-PHP services, the database server and the priced backlog. Do not run the Slowpoke Laravel package and OpenTelemetry auto-instrumentation in the same app at once, but Tideways and Slowpoke themselves do not conflict.
FAQ
Slowpoke and Tideways, asked plainly
Is Slowpoke a free, self-hosted Tideways alternative?
Does Slowpoke need a PHP extension?
Does Slowpoke support Symfony, Magento or Shopware?
What happens if the Slowpoke agent is down?
Find out what you inherited
Fifteen minutes on one server, and the first number is yours — it costs nothing to find out.