German website DerWesten has picked up on my video survey results in a blog post comparing their times with my results.

Markus Huendgen blogged about results comparing them to a breakdown of their production times for a Halloween video they created*:

1. Preparation (accreditation etc.) about 1h
2. Map (food Essen after to Bottrop) approximately 1.5 h
3. The presence on the ground (filming scenes look etc.) about 3h
4. Return (Bottrop after to Essen) about 0.5 hThus, we have already arrived at 6 hours.
5. After processing (materials processing, sift, cut) approximately 2h
6. Creating the video formats (streaming, downloading) approximately 1 h
7. Adjusting to the portal (upload, articles) about 0.5 h

Come bottom line approximately 9.5 h For three minutes of video.

*it’s a Google translation of the page

It’s a good vid (what those crazy kids will do for fun these days!) but even with improvements in the upload times, more efficiency in the editing and better setting up of contributors production times wouldn’t get much shorter.

So whilst Markus concedes that there may be benefits the reality is that this stuff takes time and takes Schweinegeld.

Roughly translated : Megabucks.

Note: Der Westen is a regional news portal by the Essen-based WAZ newspaper group. It’s a pretty impressive site with all kinds of web 2.0 goodness including geo-tagging of stories. It hasn’t been without friction though and Martin Stabe has nice post with more about Der Westen over at Press Gazette, laying out some of the issues.

So, the responses to my second little survey about video have slowed enough for me to put out some results.

I tried to expand on some of the areas from the first survey that had caused debate, in particular points raised by the Newspaper video yahoo group. So here are some of the headlines just to get people wound up:

• Video is commonly shot and edited by the same person
• Reporters are expected to shoot video and file copy on the same story
• Photographers are expected to shoot stills and video on the same story
• You should allow 4 hours to produce 1 minute of video
• There are no clearly defined roles in newsrooms for video

Production Times

The last survey asked the simple question, “On average, how long does it take to produce video“. On average this was around 2-3 hours. Coupled with an average running time of 2-3 minutes, I made a bit of sweeping statement that 1 minute of video would take 1 hour.

In this survey running times of 2-3 minutes where still the overwhelming norm. But I tried to be more specific about what production time was by splitting it in to three areas.

• Pre-production – On average how long does it take plan and prepare (storyboard, set up interviews etc.) for a video shoot?
• Production – On average how long does it take to shoot your video?
• Post-production – From the point you return from the shoot to the point it is ready to upload to the site, on average how long does it take to edit your video?

Taking the average figures for each of these sections a rule of thumb time would be more like 4 hours production for a minute of online video.

The 1hour for one minute time was a surprise to me in that it mirrors the kind of production times I would expect to see in a smooth running broadcast news set-up. The 4:1 ration brings the production time more in line with the kind of times I see in longer form broadcast productions (over 30 mins running time)

This would tally with the kind of video that people are producing:

What kind of video

Most newsrooms surveyed seem to be happy with the amount of breaking news they generate. I think this is sensible. Most editors are committed to the idea that the web is a breaking news medium (I’m not too sure about that, but that’s me) but recognise that video is not the way to break it.

Multimedia seems to be the way that people would like to develop their video output. I didn’t put too much of a definition of what multimedia actually is apart from saying ‘part of a multimedia package’. I was thinking along the lines of things like Being a black man or Reporting for Duty.

On a side note this would make me think that a lot of newsrooms need to consider a serious look at the skillset required here in terms of training and staffing.

The editing bottleneck

What is clear, both in terms of the time and type of content, is that the area that has the most impact is the editing time. I don’t see a premium on editing times a problem per se. Stories are often made in the edit suite and a lot of time is taken discussing how a story fits together – hence the off-line and on-line paradigm. But experience and some of the comments raised around this and the last survey would point more towards a lack of experience and skills in this area slowing things down. As one responder commented:

Beginning video producers can take upwards of 8 hours or longer to edit a short piece. Advanced shooters (>1 year experience) can edit in 1-4 hours.

But lack of experience is not the only factor here.

The upload tax

There is also the issue of compression/conversion to consider here. The larger percentage of post-production times over 4 hours suggests that most responders included the time it takes to convert the video in to a workable web format. As the respondent above pointed out:

The flv encoding/de-interlacing/uploading process takes an additional 1/2 – 3 hours depending on the quality desired.

So perhaps a strictly accurate ratio for actual production would be to assume a generous 3:1 of person time with a 4:1 ‘to site ratio’ - the time from idea to appearing on the site.

Who is doing it?

Video is shot and edited by the same person in 84% of newsrooms who responded . But looking at the breakdowns of who shoots and edits, there seems to be a contradiction.

Looking at the US bias is clear here with photographers dominating the production. But dedicated multimedia producers dominating editing.

In none US newsrooms the division of labour is much more even.

So what’s the reason?

Who am I right now?

One reason is that there are photographers and reporters who have dual identities. They are also digital editors or producers.

But looking through the comments I would guess that this goes one step beyond simple multiple roles. I think it’s people answering some questions based on the task they are doing at the time rather than their absolute job description. In reference to another question one responder noted

I’m actually a dedicated multimedia producer and I guess I self assign about half of my stories. I consider myself to be both a reporter and a photographer … So the answer to that one may not be exactly what you had in mind.

So I’m wondering whether people engaged in an element of multiple personality when answering all the questions in the survey. This, assuming job titles based on activity, is something that interests me and I will return to in another post.

Spinning plates

Sticking with the idea of multitasking there’s an alarming, but not surprising statistic, about the numbers expected to produce multiple versions of a story. In a majority of the newsrooms (71%), reporters would be expected to shoot video and file print copy on the same story. Photographers would be expected to shoot stills as well as video for a story in 66% of the newsrooms who responded.

If we take the headline that the same person will be shooting and editing video these figures are pretty scary. You’re going to be very busy.

But the more specific figures would suggest that reporters and photographers would produce content and hand it over to other people to edit (although I accept that in the US that is likely to be another photographer). They would be part of the chain. But the apparent contradictions would suggest that there needs to be more specific study in to the workflows in newsrooms.

Conclusions

There is a lot more information in the survey and I do think that is a better ‘finger in the wind’ than my previous effort. But I think it’s only going to make sense, and be really useful, with a lot more work around those specific workflows. (hey, I may be getting closer to actually doing some proper academic research!).

But here are two responses that I think define the boundaries of the debate. This one from the UK .

Training, training, training, training – get the picture?

And this one from the US

Our viewer numbers were way higher with photo galleries, even before soundslides but we are whole hog into video with nothing but declining numbers and tons on extra work to show for it. Our publisher thinks Utube should be our standard, I don’t know why we wasted the money on the XHA1, when we could have used much cheaper gear for this standard.

Discuss…

As I said in my last post, my little survey has kicked up quite a bit of comment.  My 1 hour of edit for 1 minute of video was especially vexing for some who feared that this may create unrealistic demands from ‘managers’.

Although I do have some figures from the survey about who is doing what I thought I needed to get a bit more depth.

So, I’m asking people to take another survey.

It’s similar to the last in that there is no personal data collected and it’s 20 multiple choice questions that I hope are short and sweet.

If you do have a spare five minutes it would be really useful. If nothing else to stop ‘managers’ making stupid decisions based on basic data.

Click Here to take survey

Update

The results of part 2 of this survey are now upResults are still trickling in but I thought I would publish a few results from my little video survey.

Some broad results:

• The average length for video is between 2 -3 minutes
• The average production time is between 2 -4 hours
• The most common camera used in newsrooms is the Cannon XH-A1
• The most common edit software in use is Final Cut pro
• Daily papers produce around 4-8 videos a week compared to 1-4 for weeklies
• Publishers with daily and weekly papers produce 2-4 videos a week
• It takes 1 hour to produce 1 minute of video

Limitations

It’s worth saying up front that the vast majority of the responses to the survey have been from the US. There is a clear bias because of this. I don’t think this limits the usefulness of the data (it was meant to be broad brush) and there are clear areas of distinction between US and the rest of the responses that go beyond statistical ‘blips’ but it would be great if more people from outside the US would take the survey – please!

Cameras

In newspaper video HDV reigns supreme and by far the most common camera in use was the Canon XH-A1. The popular HV20 plays a close second but the US bias plays a big part in those figures.

The Sony HVR-A1 has a strong showing due to its almost exclusive use in the UK and Europe.

There is a lot of use of so called Point-and-shoot cameras in the newsroom as well as some ‘lower end’ DV camcorders. Canon do really well here. Their zr800, Elura and Optura rangesseem to be popular choices.

Canon’s announcements of a scaling back on its mini-dv range may put pay to that. That may force more to move down the HDD route. A number of cameras from Sony’s HD range were a popular choice to compliment the more pricey kit. Perhaps we will see more of these now that FCP has got log and Transfer support for the AVCHD format.

One theme that was common in response to the question of what other cameras where in use, was that these lower cost cameras where specifically for reporters. The ‘high end’ stuff went to photographers. And in, what some may find, a worrying trend a few respondents noted that they were using their own kit.

Edit software

The US bias shows through again here with FCP and FCP Express the popular choice. In the UK Avid Xpress holds its own. But a common theme across the survey is the use of low and high end technology combined. Those using FCP are also using i-movie. Likewise Windows Movie Maker was common alongside Avid.

I think this says more about the platform in use than the quality of the software. In the US Macs reign supreme. PC everywhere else. However several responders listed Windows Movie maker as their only editing package. Is this disruptive technology at work or tight budgets? (I suppose given the point about people using their own kit, that one is a rhetorical question. )

Premier and Vegas have relatively low usage despite some vocal support for the packages. Again, I think it’s the platform that dictates this and I would imagine Premier making a stronger showing in future now that a Mac version is available.

Unlike the cameras there seems to be little use of anything other than the well known names. I suppose I-movie and WMM cover the bases on that one

Length and production time

The numbers seem to confirm the Comscore view of online video with majority of newspapers averaging a running length of 2-3 minutes for video.

The average production time site squarely between 2-4 hours, regardless of the size of the newsroom or technology used (until you get to less than 10), which usefully breaks down to a guide line you should expect a production time of one hour production for every minute on screen. A scary stat for planning how to do video.

Conclusion

I still have the issue of who is producing the video to look at but I think a lot of the information here is useful, if nothing else, as a snapshot of the kit to consider and the realities of the time it takes to make this stuff.

If you consider that the Canon XH-A1 retails at around £2300, FCP and a machine to run it on would be around another £2500 . Factor in the cost of losing/paying a member of staff for half-a-day to actually use the kit and you have some tough questions to ask about video.

Going down the point-and-shoot path may sweeten the financial pill in terms of outlay but the production times have no less impact on workforce and as I and many others have pointed out, giving people consumer and semi-pro kit, whilst saving your budget, often impacts on their perception of quality and professionalism.

Update: Two of the questions in the survey refer to journalists and photographers. One respondent noted that photographers are journalists too and perhaps Reporter would have been a better word. I agree but haven’t changed it for the sake of consistency of the survey. It’s a valid point though. Update

A bit of a request for help for a Monday Morning.

Are you a newspaper doing video? 

I’ve put together a little survey that asks what kind of kit you are using in your newspaper newsroom. It’s very short and doesn’t ask for any personal or professional information.  I would hope to publish the results on the blog but if you have any questions or concerns about the survey then let me know.

So, if you have five minutes then please take the Survey.