Angela Bond - an appreciation

An old friend and long-time mentor, Angela Bond, died last week. She changed Radio.

Angela with Kenny Everett back in the day

Angela Bond was one of a kind. An old school radio pro; you couldn’t find better. 

Her worldwide career spanned singing, presenting, producing (Kenny Everett and Ed Stewart, among others, at Radio 1), slugging her way through BBC seventies and eighties bureaucracy, and then ushering in some of the most seismic changes to hit UK music radio, as the first person to rep Selector, the original music scheduling software, for the UK. 

While people outside radio - and many inside - may well not have even heard of Angela, it's almost impossible to calculate, let alone explain, the impact she had.

Case Histories and conclusions

The Track Record page here and on my website has turned out to be of the most consistently visited pages. To build on this, I’ve just gussied my website with some of the more interesting Selector case histories. If you’re at all interested in music scheduling in radio, there may be stuff to glean; feel free to help yourself. 

The case histories are, in chronological order:

BRMB-FM / Xtra-AM: the highs and lows of introducing computer scheduling. Very early 90s.
BBC Radio 2: putting in a multi-genre library, along with the BBC's very first digital playout rig.
Swedish Radio P4: developing client skills in hugely varied markets.
RTÉ lyric fm: working collaboratively with passionately involved production staff to build not so much a database as a knowledge base. The polar opposite of most implementations.
UTV Radio: upskilling staff and debugging inherited nightmare scheduling conflicts.
Coast 106: swimming successfully against the UK radio stream with a larger than normal library.

In this diverse range of situations, there are some common threads...

First : dialogue, up and down the chain of command,  is good. In fact, in my view it’s not so much good as essential. While many radio stations implement a rigid schedule from above, normally for ease and simplicity of management, some of the best ideas and approaches evolve from engaging with the staff who work with the system. Nobody is right all the time. If there is a conflict, either with content or with programming, it’s often very useful to examine that conflict in minute detail, so see if there is a better way to do things. Best to leave your ego at the door, though. I won’t name the middle manager who loved the idea of challenging his boss on music issues, but hated the idea of talking to his own staff about those same issues.

Second: We’re in the era of tiny databases. In the US, they’re now talking about cutting down from 150 to 50 songs. However, almost all the above case histories show ratings success allied to larger libraries. 

Third: A note to managers: large databases can be a pain to keep tidy. And talking to your staff about programming issues can be a hassle you could do without. Things can get emotive. It can eat into your time, and not everybody has that luxury. But I suggest that if you actually care about what you do, you’ll benefit from putting in that time. Radio is still full of passionate people. You’ll get the best out of those people if you meet them halfway, here and everywhere else where ideas can be shared and debated.

Own your music: Selector (and other scheduling systems) tip 4

Number four in a series of tips for Selector (and other scheduling system) users.

When I talk about ownership, I’m talking about really getting hold of the material you’re going to play. Simply put, if you have no real idea of what’s important in your library, or why, you’re going to sound messy on air. A  big, deep repertoire is a wonderful thing. But it's absolutely no use to you, unless you know what you’re going to do with that repertoire. If you think about it; you win. If someone else does the thinking, or worse, nobody does the thinking, you lose.
 

An example: A small-scale station was setting up at the end of last year. They had no money for a library. So staff lent their own CD collections to copy to the station server. This is not at all unusual. They happily ripped everything they had: complete albums, dozens of them. This rapidly built up a library of several thousand cuts. However, the station’s system just happened to be off-line… so, no CDDB/Gracenote, and no artist or title recognition. Giving a database full of Song1, Song2, and so on, performed by Artist1, Artist2… and so on.  And, because they were ripping complete albums, including hits packages, that database was also full of duplicates. That’s a nightmare scenario, one which will take far more time to correct than it would have taken to think about core material in the first place.

I’ve seen this happen too on courses I teach at college radio – I do a couple a year. Hundreds of tracks get dumped into memory, just to get something to work with. That’s not building a library.

Here’s why. Take a look at your Ipod. Chances are you’ve done much the same thing, especially if you’ve got a lot of memory and you want to have a BIG library to brag about. And how is that huge choice working for you? Have you taken the time to go through everything in iTunes and prioritise every single track in to favourite and less favourite? I'll bet you haven’t. What about deleting stuff? Tricky. I bet your shuffle play is… random.

Here’s the thing. Even through we’re in the era of single song downloads – a healthy development - we’re drowning in easy availability. We don’t value songs for themselves anymore - that’s so last century. Music, like so much else, has become commoditised, and it’s not healthy.

I believe that each song needs to be looked at hard by any radio programmer, before being put forward for scheduling. Don’t misunderstand me: I am not arguing for 100-song playlists like they have at a lot of stations in the US these days. There’s no reason not to go ahead and build up a huge library. Just make sure, before you do, that there’s a good reason to add each song in the first place (somehow, ‘I just like It’ isn’t quite enough).  Thinking about your stuff before you add it also helps you to build a coherent structure for your scheduling system. Core songs by core artists – no matter what your format – is a great place to start. You can work it out from there.

And it might be wise to make sure you’re online when you're ripping :-).

If this has been useful, pass it on to friends and colleagues. It’s on me. If you'd like more, on a 1 to 1 basis, reply to me through the blog, or email me via the website (link at left under Work-related).

My Selector Coloring book: Selector Tips 3

This topic mainly applies to Selector version 15 lovers. You can do a bit of color tweakage in Version 12 (that’s the old, ugly, but gorgeously stable and well thought out dos version), but it’s really the windows version of Selector that lets you play with looks, fonts and colors. There’s lots of configuration potential in GSelector, too, but that hasn’t fully rolled out worldwide, and I’m quite sure that these remarks will also apply just as much to other scheduling engines. Bottom line? Customise away, but you should avoid the explosion in the paint factory effect at all costs.

Why is it that contemporary software design is so…. uniform? There’s a very good reason. Computer screens can display a LOT of information. Color is a great help is highlighting areas of concern, and you can often set conditions in your software package – not just your scheduling engine – which will throw a focus on an area of interest, by using a specific color.

But if you make things too busy, your brain has to work a lot harder to take it all in. If the screen is just too busy, you tend to jump past all this information.

Now let’s consider the Editor screen. That’s the one that displays your schedule, or running order. You’ve got to review an entire day of output – that's at least 24 screens, maybe much more.  If the entire screen is a maze of color, you’re going to have a hard time concentrating as hard as you need to for your editing job… which means you might let something slip past… which means the output might sound lousy.

So that’s why I suggest you go easy on the color.  If you like to differentiate between different categories onscreen, that’s fine – but try using shades of the same color, rather than violently clashing and distinct colors. Leave the fireworks for the emergency conditions: schedule failures and the like. And think about whether you need to apply color to the entire page – you can restrict it to one or two fields if you prefer, leaving the rest of the window more uniform.

Above all, make it easy on yourself, so you can make those critical editorial decisions: getting the mix right is way more important than having a pretty screen display that your audience doesn’t know or care about.

If this has been useful, pass it on to friends and colleagues. It’s on me. If you'd like more on a 1 to 1 basis, reply to me through the blog, or email me via the website (link at left under Work-related).

Inheriting a Selector scheduling setup: Selector Tips 2

Most of us in Radio inherit a Selector rig when we join a station, or when the guy before us moves on. Nowadays, you very rarely build up a system, your way, from scratch. So you really need to work out how the guy before you went about his business, before changing things. If you don't, then you’re in for a world of pain working out why the rule settings don’t seem to work right anymore. If the guy before you was obsessive, you better be too. Sorry about that. The basic principles here don't just apply to Selector - they work in any system. I’ve got a how-to tip to help after the jump.

The devil is always in the detail.
A great way to zero in on any aspect of your library is to go to the Browse Window. Let’s say you want to make sure your Artist list is accurate. You really need this. Selector isn’t Word, and it won’t compensate for spelling errors.

Click on the Artist heading.
Work down the list of songs, starting from the top.
Some errors might be right at the very top – like songs with no artist.
Look for anomalies. Are songs suspiciously absent? Maybe they’re listed elsewhere under a mis-spelled artist name.

And don’t forget the famous text string issue – Selector normally sorts by the first letter of the last piece of text (known in geekland as a text string) in the field. So it would see Take That and Take_That as two different artists.

You can repeat this basic exercise for anything that is key in your scheduling rules, but of course there are some quite complex areas to make decisions on, above and beyond factual issues like getting the Artist names right. I'll post some tips in dealing with the more subjective areas soon.

Sorted? Good. Now make a note in Outlook to do this all over again in two months.

If this has been useful - pass it on. If you'd like more on a 1 to 1 basis, reply to me through the blog, or email me via the website (link at left under Work-related).

Coast 106: I love a success story

And this is a success story. I’ve been working as a consultant with a lovely and thoroughly listenable radio station in Southampton. They’re called Coast 106, and I’ve been visiting them on and off since late 2008. In this quarter’s Rajar audience survey, they posted all time high figures. In fact it’s been a pretty solid story of growth from an admittedly low start point, when they took the franchise over from the previous owners.

It’s pretty hard to rebrand a station, and then relaunch it in the teeth of a howling recession at any time. Harder still to post consistent growth figures - doubling reach and nearly trebling market share over an 18 month period. So what’s their secret?.

How do I know my Selector Mood Codes are right? Selector Tip 1

New feature. I’ve been a Selector wrangler for over half my career; it’s partly how I make my living.

I’m going to post a regular series of tips. Most of these cover editorial approach, which I think is an area that gets left behind. Some tips will apply to other scheduling systems. If you like what you see, or if you know someone who might, pass them on. If you really like what you see, get in touch by replying directly to this post (bottom of the post, below) or through the website link in the Work-Related pane at left.

Tip 1 is about finding your Centre Of Gravity - the midpoint of your Mood or Energy values.

The overall balance of station sound depends hugely on how you classify songs. Typically stations code their songs for Mood or Energy (occasionally both), on a scale of 1 to 5. 1 is catatonic, 5 is extremely in your face.

This is, naturally, a very subjective area. So it really helps to get a very clear idea of what the mid-ground Mood value is for your station. That’s what I mean by Centre of Gravity.

Here’s a recipe:
- Open up a list of all your active songs.
- If you have coded songs for Mood or Energy, sort them by this field.
- Now look in the middle area to find, say, three songs that are absolutely bang in the middle of the Mood range for your station. That's your benchmark.
- Work out from there. Work back up the list, and measure every song against your new standard.
- Then go down the list the other way. I guarantee you’ll find some surprises.
- If you have a big library, break it down into Categories first. This can be heavy going.
- If you haven’t set values at all - shame on you - the process is still the same. Find three benchmark songs, and work out from there.
- By the way, this is hard work. Fatigue can set in. You probably won't get it done in one session. Take your time to get it right.

After you've done that initial re-evaluation, do some analysis work to see if you’ve set the Mood or Energy rules up right for what you’ve actually got. That’s a tip for another day.

Who's listening to what?

I spent some time yesterday at Absolute Radio's studios in London. My main purpose was to take a closer look at one of their new online products, comparemyradio. This is a very cute piece of software, and it has caused no end of fuss in the very closed hothouse world of UK Radio. Basically, this software tracks plays of songs on a number of major stations – most of the big ones, in fact – and lets you compare who plays what, how often, and where the overlaps are. It’s open to anyone to play with, and it throws up some very interesting perspectives.

Full disclosure: I author help for a heavyweight professional media analysis software package called Media Monitors, a sister company to RCS, the Selector people, for whom I also write help packages. So I had more than a pure radio interest: I wanted to see what the One Golden Square Labs boys were up to, what their analysis kit could do… and, above all, exactly why they chose to share all this with you and me.

Absolute were the soul of hospitality, in keeping with their mission statement of openness and accountability, which I applaud. I’m all for exchanging ideas and comparing notes. Personally, I'm not threatened by the idea of sharing what I know. Those that do feel this way, in my opinion, lose out, and wind up hobbling their own creativity. That said, I didn’t quite get all the answers I was looking for, but I’ll come back to that.

Comparemyradio went public last week and instantly kicked up a fuss. Bauer Media’s Kiss-fm promptly disabled their 'now playing' site feed, which is where the comparemyradio software scrapes its data from, arguing that their daytime output does not reflect the full range of output from their specialist shows, which presumably do not feed their data out in such detail. When I looked this morning, BRMB had also lifted up its skirts and (note this page may have changed again if you check it) removed its output from Absolute’s grasp. Other industry operators have publicly expressed their irritation.

This is being just a trifle precious. A station’s output is public. Once broadcast, it’s published, it’s out there. Anybody who wants to take the time and trouble can study the output, analyse it, and draw their own conclusions. What comparemyradio has done is simply make this a lot easier. However, some stations don’t exactly look that great under this new bright light. So it’s fair to say that Absolute are being a mite provocative. The most provocative tool of all is something they call the variety index. This charts the ratio of lightly played material as a proportion of overall output. Clearly, the greater number of songs which get unique plays over a given period, the more genuine variety the station can claim to be offering. To put it another way: if you play 300 songs a day, and 150 of those are unique plays, you’ve got a lot of variety. But if you play ten songs six times a day, another 30 songs three times a day, and 60 more twice a day, there’s only 30 single play slots left in your day, so you’ve got less room to mix it up.

I really wanted to do a comparison on all West Midlands stations, as this blog focusses on this area. I’ve have loved to have BRMB’s figures, but, as mentioned, they’ve blocked this now (boo! hiss! shame!). But Heart West Midlands has a variety index of 11%, not entirely encouraging in a station that claims to offer more music variety. Smooth London, whose output is uncannily close that that at Smooth Birmingham, can offer a respectable 26%. But scooting over to Radio 2, the big beast in the radio jungle, we see a huge variety index of 60%, which must reflect their detailed and admirable specialist show output, as well as their large (by today’s standards) active daytime library. It presents them in a frankly golden light, but then, look at their listening figures. And what about Absolute themselves, purveyors of this provocative package? Well, their flagship service offers a (not at all bad, but could do better in my view) figure of 20%.

So what are we to make of all this data? Well, for people like me, it’s a lot of fun, and it’s a useful new rough and ready tool to get a feel for certain areas. So I’m quite pleased it’s there. It’s certainly stirred up a hornet’s nest of resentment in the industry, which I can happily live with. There are questions as to the validity of the data and the accuracy of the source material that is scraped to analyse in the first place, but Absolute do acknowledge this. I also heard dark mutterings yesterday about the use of other public streaming sites’ software to do some of the heavy lifting in the analysis area, and that worries me – the more third party involvement, the less trustworthy the data tends to be. But I have no way of verifying this.

The primary beneficiary of all this has to be, of course, Absolute Radio. They have offered their users something new, which might just be a novelty. Or it might be something to raise the company’s profile – chalk up a success there – and keep users on site to explore some of their other products. So in terms of building relationships with listeners while dissing the opposition, it’s pretty cute and Imaginative.

A closing note: Anthony Abott, the courteous and affable Absolute Radio webmaster who spent time with me, and to whom I extend my thanks, checked this blog out before I arrived. He in turn was interested in my championing of local bands and my mantra that local radio benefits from exposing the best of its local talent. Now, this is something that makes no sense whatsoever from a London or a national perspective, so I can see why he was curious. I’m sure there are millions of interesting music scenes in London that aren’t getting the exposure they deserve – that’s the problem with radio in big cities: it's too damn conservative because it's too damn competitive. New York is the same, which is even more tragic.

But this really does give operators outside of the smoke a big advantage, where cities and regions have much stronger senses of local identity. And how does this tie in with what I’ve been discussing? Simple. Forget about variety indexes just for a moment. Forget that hugely depressing overlap between stations. Look instead at the USP for your market. This is something the big boys can’t ever do. If nothing else, comparemyradio has shone a dirty great light into this area. It’s up to local stations to exploit what they can learn from it. If that includes bumping up a variety index by championing bands that you know for certain matter to your audience, then it’s all good.