any2.video vs Recording It In House: The Tools Are Free, the Engineer's Afternoon Is Not
Recording videos yourself is the cheapest way to get your first three, and for many companies it is the right decision. It breaks down once coverage starts to matter, because what limits you is the one person who knows the product well enough to demonstrate it.
What each option proves
| Question | In house | any2.video |
|---|---|---|
| Real cost | An engineer's afternoon for every video. | One sentence of brief and a review. |
| First video | This week, if someone clears the time. | Two days. |
| Video twenty | It rarely happens, because in-house programmes stop before it. | A batch. |
| Environment | Usually someone's laptop, with their tenant and their data. | A clean environment built for the run. |
| Version tracking | Whatever discipline you impose on yourselves. | A version tag on every video, re-run on release. |
| Failures | Edited out, because they feel like mistakes. | Kept in, because people search for them. |
When recording in-house is the right choice
- You need one or two videos, and someone is willing to make them.
- The audience is internal, or a single customer who already has the context.
- The product changes slowly enough that a recording stays accurate for a year.
- You want to see what your own tasks look like on camera before buying anything, which we recommend.
When it is not
- The list runs to two hundred tasks, and the person who can record them is also the person closing deals.
- The videos sit on evaluation pages, and nobody is responsible for keeping them current.
- You need the same task in three languages.
- The recordings show a laptop with customer names on it, the most common reason an in-house library never gets published.
Start in house
The advice on this site has been the same since launch, which is to make the first video yourself. Pick the task your support team explains most often, record it, publish it and see what happens. You will learn more from that than from any vendor conversation, this one included, and you will know whether the demand is real before spending anything.
How to start video marketing at a software company is a guide to doing exactly that.
Where it breaks down
In-house programmes rarely fail because of quality. They fail because the numbers do not work. A mid-sized product has hundreds of tasks a user might need, across platforms and languages, and it ships again next month. One person recording between other work produces four videos a year, and three of them are out of date by the time the fourth is done.
The second problem is less obvious. The person recording already knows the answer, so they never run into the wall a new user hits, and the recording ends up answering a question nobody asked. The useful footage is the 403 on the first attempt, and someone who knows the product never gets one.
The mix most teams settle on
Support records a rough answer the day a ticket arrives, because speed matters more than polish. Tasks that keep coming back get made properly, verified and versioned, and go on the pages prospects read. Teams end up doing both.
Questions
Should we just do this ourselves?
For the first two or three videos, probably yes, and we will tell you so. The tools are free, and you learn what your tasks look like on camera. Revisit the question once the list is too long for one person to keep up with.
What does it actually cost us to record in house?
The tools cost nothing. What you pay for is the time of the few people who can operate the product, and they are usually the same people handling support escalations and pre-sales. That cost is what ends most in-house programmes.
Can you work from recordings we already made?
We would rather run the task cleanly than patch an old recording, because the version tag and the evidence are what make the video worth having. Send the old one as a brief and we will tell you what a verified version would change.
The fastest way to settle this is one task, done both ways. The first sample is free.
$ get-sample → $ cd /library