Ovum is floating -just about

As we announced before, Ovum has now gone through their IPO on the 10th at 190 pence. Their stock shot up to 219 on the 12th and went back to 198 on the 24th in small volumes.

Some in the industry have been speculating on what may happen next:

• A mass exodus of suddenly enriched analysts?
• An acquisition?
• Aggressive development plans, for instance in consulting?

Read also:

• Ovum to IPO, Forrester loses sales VP
• Has Forrester missed a golden opportunity?
  
• Making it up While Drinking
• Ovum seeks intelligence head, Gartner Watch in quest of integrity
• Weekly digest
• Ovum breaks the iPod cellphone embargo?
  
• Ovum makes a move!
• Furl'ing: Ovum, Aberdeen and Gartner in the blogs
• Is Ovum ambitious?
• 1+1=1.2?

NEW: AR 101 feed

In response to the success of our popular AR 101 series, we've introduced a specific feed, with a number of subscription options below. Note that any input and suggestions for upcoming topics is welcome.


RSS (Atom + XML feed)

Add to your MyYahoo!

Add to your Bloglines

Receive ARmadgeddon's AR 101 Series by email:

AR 101 series: briefing analysts

James, Jon and Andy kindly provided quite a lot of (free) advice on how to brief IT Analysts.

Here are six simple steps to get your briefing right (tell us if we forget anything guys).

1. Synch briefings with research agendas
A good briefing should be prepared before to make sure that the information given to analysts match their research agendas and interests. This is of course easier done for one-on-one briefings (some simple filtering questions at the start help). For one-to-many briefings (not the preferred briefing for most analysts but often the most practical compromise to update analysts on announcements in a short timeframe) it is good practice to send 2-3 questions to key analysts before the call.

Read AR 101 series: the research process for more. Jon made the following remark:

• Take the time to understand research programs and assignments of the analyst firms you have in your sights, and use this information to make the correct people available to support the analysts that you meet. For example, if a firm is conducting a market analysis study, there is little point in providing a technical expert at a briefing; similarly, a brand manager will be of only limited use if a product comparison is taking place. Either of these situations may lead to the worst-case scenario of not being mentioned at all.

2. Don't give analysts a dog's dinner
Bear in mind that you may be talking to a diverse audience. Analysts fall into four categories: RAS analysts, market watchers, consulting analysts and sell-side analysts.

So, as Jon says in this great post, you need to "Make Briefings Worth It":

• "At the best of times, briefings can be dull for analysts as much as for vendors. This is often down to the fact that briefing sessions are inappropriate, badly planned or conducted."

He suggests that a good preparation is paramount:

• Only use source information that is relevant to the briefing at hand. Don’t waste anyone’s time in briefings, by slogging through irrelevant presentations that have been picked off the shelf
• Customize your message before the briefing

Analysts usually want to hear the following:

• Corporate strategy
• What are you announcing? (high-level messages AND product details)
• How is that going to fit with the strategy and go-to-market model
  
• How are you going to compete?
• How does this fit with your announcement roadmap?

And be flexible to respond to analysts prompts during briefings, do not hesitate to skip the introduction and go directly to the point. A common mistake is to explain granny how to suck eggs, for instance by lecturing numbers analysts (say from IDC) on market stats (say from Gartner/Dataquest). Andy illustrates this graphically in his post:

• "That mistake is mostly made by smaller companies, where some VP or Director of Marketing is doing his (or her) run-of-the-mill pitch that would be equally used in any sales opportunity. So they go on and on about stuff that we have heard a million times already, have zero value to an analyst, and simply waste time. It's particularly annoying if those briefings are conducted via WebEx, because the analyst cannot [...] tell the vendor to skip to slide 32 [...]"
• Vendors that do it very well [...] spend two or three minutes on chit-chat [...], then run through a few figures [...] and not more than 10 minutes after the call starts jump right into the product update, demo, or discussion.

Read also the comments Vinnie, Dean and Jon have left here on the subject. We had the following bottom-line in the same post:

• ARmadgeddon is against unnecessary analyst cruelty and agrees that vendors could do a better job at scoping briefings to better address analysts needs.

Finally, read also: AR 101 Series: Don't use a sales presentation with analysts

3. The best briefings are interactive
Do allow time for analyst feedback during briefings ; bearing in mind that there must be something in for the analyst -the analyst should not feel like she/he is giving out all his IP for free. Bear in mind that analysts need to make a living and allow budget for buying reprints or getting the analyst under retainer. When the interaction is good, your company will gain a lot.Do not let short-term tactics waste this potential by making sure that your briefings are win-wins.

As Jon writes:

• If you are able to provide an analyst and his or her company with timely, tangible value, the analyst is more likely to have something to say about your company and your products.

James made similar comments in his post Redmonk: how to brief analysts):

• So, an analyst briefing is (should) a two-way conversation (otherwise it's called a press conference).

4. Don't forget NDA's
Analysts are interested first and foremost by future strategies and roadmaps. Do balance this requirement with the need to keep your job (and those of your executives) by being careful to flag what is presented to analysts under a non-disclosure agreement. Always keep the required paperwork at hand.

Read also Ovum breaks the iPod cellphone embargo?

5. Do a dry run
As we've said before: NOBODY should be talking to analysts without going through a good AR training first. This should be a corporate policy. If you don't have the skills or the credibility to deliver it in house, get some professional help. For instance from Duncan or David (to name only some who linked to us), if you pick-up someone else, make sure that they know the European market (not like KGC).

Furthermore, AR professionals should rehearse unless they're fully confident in the speaker's capabilities. This will not make them look appear as PITA's but rather as pros who need to make sure they avoid un-necessary risks.

5. Follow-up
Finally, do make sure that AR does the follow-up and stays in the loop -possibly by recommending that executives don't give out business cards. This should not annoy the analyst if you're responsive and has the benefit of being in a position to schedule additional briefings when required and stay abreast of the analyst agenda. This brings us back to step 1!


Additional bedtime reading:

• ARmadgeddon: Ovum breaks the iPod cellphone embargo?
  
• bit blue blog: Vendor briefings: Do more with less :: Andreas Bitterer ::
• ARmadgeddon: [Monkchips:] No Time Wasters Please: On briefing industry analysts
• Jon Collins:Nothing to Declare » IT Analysis In A Cold Climate
• Comments on the following ARmadgeddon posts: Gartner limits vendor briefings to 30 minutes and Borg imposes 30mn briefing limit

[deal architect:] Credibility of Analysts

Vinnie has a special talent for finding the sweet spot and sharp comments.

In this post, [deal architect:] Credibility of Analysts, he comments on the Borg IP police: the Gartner Ombudsman blog.

They posted a lame rebuttal to the Information Week article on Analyst Credibility. The comments on that post (the Borg must be congratulated, yes congratulated, for having a real blog) are worth a read.

Carter Lusher, the HP (HPQ) AR head argues for greater separation and transparency.

Chris Carter (?) asks two penetrating questions, only to find deafening silence. Follow-on through to Vinnie's blog for more....

The Borg AR call

We were hoping that Joe, Duncan or Dave would write a take on Wednesday's Gartner call as it was really quite dull... The only really funny thing was that Pamela Miranda referred to the Miranda act, so she must have read this Gartner Watch post.

Laura McLellan, Gartner Research VP was first to speak. She covers vendor (IT providers) marketing strategies research. She started to say Gartner was not going to start to analyse AR (as the Oracle AR VP pointed out, this would be a conflict of interest) and compete with KGC.

She had an interesting take on analyst taxonomy, depicting it as an inverted triangle: on top analyst which are end users focussed (advise buyers), in the middle market watchers (who look at forecasts and trends) and on the bottom (tip) the vendor focussed analysts (advise sellers). Quite why the triangle is inverted, we don't know -there may be more user facing analysts in Gartner but in the overall analyst landscape the vendor facing analysts outnumber them by far. ARmadgeddon will be publishing an official Analyst Taxonomy soon. She then said that you need to adapt your approach to the kind of analysts. Not exactly new news but maybe a useful reminder.

She then further continued to explain grandmas (AR professionals listening into the call) how to suck eggs: one needs to segment the communication type to audience. Laura also plans to research which AR comms are effective or not and why. That would indeed make up for an interesting and more specific call.

She gave a quick round up on what analysts love most: face time, AR contact list (inc. responsibilities), announcement advance notice / pre-briefing and email opt-out.

She gave some insight on analyst psychology (gives a new meaning to analysing the analysts): they are motivated by influence and knowledge (explains the ego side) and turned off by community and reprocity (not sure what James will have to say on this).

These other points were on her charts:

• Semi-annual tours with topic managers to discuss the “bigger picture”
• Portal to find “basics” without bothering AR person
• On-line presentations and [important] transcripts
• Upcoming events calendar
• Annual executive sessions [access to the “big cheeses”]
• Opt out “push” e-mails with the news
• Access to your customer presentations

Probably worth a reminder indeed. She finished by saying what makes a good AR person: being a good relationship builder. What does the R in AR stand for again?

She gave the following good and bad for AR managers:

• GOOD: Enabler vs. roadblock, Proactive, Accountable, Responsive, Truthful, Trustworthy, Empowered, Persistent, Follows up, Management confidante, Value creator, Helpful, Relationship builder
• BAD: Defensive & suspicious, Gatekeeper (analyst = enemy), Unresponsive, Self-important (does not mix well with analysts?), Territorial, Siloed (not like Gartner then?), Roadblock, Inconsistent, Inaccurate, Don’t understand own organization, Don’t understand how analysts work, Not influential in organization supported

She finally hinted that Gartner might create a research role to look at vendor marketing.

The call then moved on to Pamela Miranda, Head of Vendor Briefings who spoke about briefings. She did not really cover the points that were on her agenda:

• How to seek analyst attendance at your events
• What’s the latest on vendor-paid travel
• Focus: The most frequently misunderstood parts of the process

Joe commented on the scheduling process here: Gartner's briefing on their briefing policy (Gartner Watch). As we've commented before, this process is lengthy and time consuming. One of the reasons that was

Read also:

• Gartner limits vendor briefings to 30 minutes
• Borg imposes 30mn briefing limit
• The replay and charts might be some day available there

AR 101 series: selling the value of AR

As we previously wrote, AR is NOT an outbound tactic: briefing analysts and expecting them to write something is called PR and does only harness a small part of the value AR can deliver. However, few AR people feel confident selling the value of AR back to other stakeholders, such as sales and marketing.

Here are a simple 5 steps guide to getting it right.

Step 1: establishing AR
AR often grows out of media relations and tend to have communication as a reporting line (this seems to be the prevalent model for small organisations). AR practitioners should seek to clearly differentiate their audience (read AR 101: Analyst vs. press) as it will impact management and stakeholders expectations.
In this post (AR 101: Measuring Analyst Relations), we lay out three models for AR:

• Outbound: pushing information out to analysts in the hope of generating reports
  
• Transactional: pulling targeted analysts into a relationship with the firm
  
• Insight: primarily using analysts to develop internal understanding and direction
  

AR professional should strive towards establishing an insight-based AR practice and align with sales and marketing rather than with comms. See Duncan's tips here.

Step 2: setting goals and measurements
Thanks to Pavlov, we know that rewards conditions behaviours: it is crucial to align AR with sales and marketing goals. In this post (AR 101: Measuring Analyst Relations), we give the following recommendations:

• AR's primary goal should be create a positive external environment for business
• Focus and attention should be given to setting up balanced metrics to measure AR not only on raw clippings and quotes but also by weighting them to reflect the importance of the analyst, by surveying all analysts and through independent perception audits.
• AR need long term objectives and can contribute positively to developing corporate strategies.

Read also ANALYST EQUITY: Share of Voice: Useful, Often Vital.

Step 3: impacting the strategy
Well executed, AR can leverage the analysts privileged position and wealth of knowledge as a competitive advantage. Do use analysts to get a sanity check and/or coach execs, balancing independent analysts (they offer better value) from those having customer insight. Do schedule message testing sessions way ahead of launches (under NDA). Do allow time for analyst feedback during briefings (bearing in mind that there must be something in for the analyst, like buying reprints or getting the analyst under retainer). Do not let short-term tactics waste this potential.

Step 4: Shaping the agenda
Having engaged proactively with analysts should give your company clues about trends you might not have spotted before. This not only gives your execs the chance of fine tuning their strategy and messages, but armed with this knowledge you should be able to also synch briefings and announcement with the analysts' research agenda. Read also AR 101 series: the research process.

Step 5: delivering to sales and marketing
There are multiple ways for AR to help marketing. Do start by making sure that AR tactics are included into the marketing plan, it should also provide for analyst deliverables (such as speaking engagement, research projects, etc...) Do plan AR tactics way ahead of product launches.

Do spend some time to explain how analysts come in the sales cycle and do raise awareness with the sales community so that they come back to AR when they hear of an analyst being consulted by the prospect or client. If you have followed step 4, you should be able to know where the analyst is coming from and what should her/his objections. You should thus be in a position to balance this with either positive research (from another firm or sometimes from the same firm) or to help sales refocusing the discussion. Remember that sales always get the kudos for wins and will try to deflect the blame for losses.
Read also Reacting to an analyst attack and Advisory analysts do impact vendor sales and make/break products.


Good selling!

Borg softens 30mn rule

In an email today, the Gartner Vendor Relations (briefing police) and Ombudsman (IP police) appear to have read ARmadgeddon's posts on briefing length:

• the 30 mn limit is now only a proposal
• they recognise the vendor briefing request process is f****d-up, as vendors commented it leads to intolerable delays
  
• analysts can extend the time if they deem it "required to obtain maximum value from the briefing"

ARmadgeddon's comment: recognising a problem is a first step to correct it, we will be monitoring Gartner's briefing process closely in the coming months.

In the meantime, there has been a lot of interesting comments on what briefings are and how long they should be, mainly here, there and there:

• We agree with the analysts that vendors need to do a better job at briefings and to avoid death-by-powerpoint. Vinnie quite rightly says that briefings should be more focused ; that pitching to analysts is quite easier than pitching to the Wall-Mart procurement team.
  
• There seem to be a consensus that 30 mn are okay for a quick update on a specific offering but does not allow enough time for wide-area-analysts (WAA, as opposed to Narrow Silo Analysts)
• Dale and others however said that restricting briefings to a mere half an hour would eliminate the discussion. James argued that this was precisely what the Borg intended ; that it is an opportunity for independent analysts.

We look forward to today's call where the Borg will share its recommendations for AR professionals, and more.

Borg imposes 30mn briefing limit

So it looks like the Borg is about to limit briefings duration: Gartner limits vendor briefings to 30 minutes.

Vinnie, Dean and Jon have good comments on this, the two sides of the argument are:

Many vendors impose a death-by-powerpoint sentence to analysts with 3 hours and 54 charts briefing. This is BAD practice (see links below). A half an hour session should thus force vendors go straight to the point an be more productive.

Analyst briefings should be interactive ; 1800 seconds does not allow much of a two-way conversation.

ARmadgeddon is against unnecessary analyst cruelty and agrees that vendors could do a better job at scoping briefings to better address analysts needs. However, 30 mn is too short to allow for an interactive conversation. The fact that this new rule comes from the Borg is not a surprise as Gartner analysts tend to be more quiet and provide significantly less feedback than independent analysts during non-paid briefings. As a side note, arranging briefings through vendor relations takes Analyst Relations Managers anywhere from 1 hour to 2 days...

POJ's wrote a nice take on this on GartnerWatch: You have the right to...
His point that Gartner analysts would struggle to explain anything in less than 40 mn is spot on.


More from ARmadgeddon on analyst briefings:

AR 101: Jon Collins on briefing analysts

[Monkchips:] No Time Wasters Please: On briefing industry analysts

Redmonk: how to brief analysts

Will the Borg be dis-intermediated?

Interesting and thought provoking post by former analyst Dana Gardner on why blogs and the information abundance is a threat to Gartner:

Self-IT analysis is the wave of the future (Dana Gardner, ZDNet.com)

His analysis is quite similar to Jame's (read this).

There's of course the reverse argument: too much information kills information, there's a role for someone to package it for users and make money in the process (read this post on Vinnie's blog).

We however remain sceptical about Gartner's consulting business (would be nice to hear from the chaps at Accenture, Deloitte, Cap, PWC, IBM, E&Y if they ever lost a deal to Gartner?)

The Governor, ancient Iraq and Gartner

Like a modern David, James has decided to take on the Borg Goliath:

• James Governor's MonkChips: Why Open Source Analysis Will Grow: Learning from The Sumerians

In this post (and before), he argues the cuneiform writing is on the wall for Gartner Borg:

• Why would vendors continue to pay a ransom to a gatekeeper to enterprise purchasing?
• They're under pressure to disclose their "methodologies" and clients
• He quotes Jonathan Schwartz, COO (SUNW), as not being happy with Gartner (IT)
  

His predication is that open source research will form a new ecosystem and make Gartner obsolete, just like an old mainframe. Vinnie says it's like Gulliver and the Lilliputs.

Joe has already posted an analysis on Gartner Watch:

• (From: Monkchips) Gartner is like a mainframe in 1979. Ripe for deconstruction.

ARmadgeddon's take: commoditisation of research is an inescapable trend and presents a serious threat to traditional content-based IT Analyst Research firms such as Gartner. However, open source research faces challenges before it can be seen as a credible alternative:

• Reputation/credibility lies with individual analysts or firms (and not with the "ecosystem")
  **An AR professional emailed us that the main issue with "one-man-boutiques" for them was the lack of branding/the confusion in the marketplace.
  
• The output of independent analysts varies considerably as it is not like a single source code (although the emergence of aggregators such as IT-Analysis.com is a major step forward). **As Richard Stiennon puts it in the comments, "open source research would imply a wikipedia like collaboration on market research, [...] the results however would be FREE to use by anyone." (see comments for more)**
  
• Consulting to users remains the exception for independent analysts and it is not proven can the open source model can be financially viable for them. **Vendors value this user insight more than everything else, this is however currently the priviledge of Gartner, Forrester, Ovum, etc....**

**27/2 edits**

Odd bods

Who are the oddest analysts around?

Cast your vote there:
James Governor's MonkChips: Who is the most off the wall industry analyst in the business?

Vinnie on the Gartner Offshore Application Services Magic Quadrant

Interesting take from Vinnie on this Borg Manic Quadrant:
deal architect : Gartner MQ on Offshore Application Services

Is there an afterlife for analysts?

Duncan asked yesterday Why vendors hire analysts.

The underlying question lies around analysts career development.

You see, analysts do have somewhat atypical jobs -which may explain why they are often peculiar themselves. They come from various backgrounds, either from IT departments, IT vendors (most frequent case) and less frequently from resellers or other channel partners, often ticking well over 15 years of experience.
Their analyst job is often fascinating, both intellectually and for they have a privileged industry vantage point -sitting between so many constituencies (vendors, users, channel, etc...)

So it's not a surprise if analysts tend to stick around and progress naturally within research firms, however even in Gartner there can be only so many VP's... Creating their own consultancy can then appear as an attractive option. Duncan describes these market dynamics here: From the 'water cycle' to the 'analyst cycle'.

Another option is to move into AR. As James and Duncan point out, running AR for an IT vendor is sometimes a rather brutal landing. We also had some comments from analysts on being dealt with by ex-analysts (this may cause some clash of egos), on the other hand ex-analysts are well positionned to sell the value of AR to the business. Another option can be to work for or create an AR consultancy (KGC, LighthouseAR are run by former analysts) -but it's quite a niche market...

Frost demands cash for press release

In ARmadgeddon: AR 101: Analysts and press quotes
we explained that analysts don't charge for quotes. However, itis worth adding a postscript: one reader has pointed out that there's one case where vendors do pay for a press release. This is what she writes:

"I just wanted make a comment say that you have to be either green or naive to take this idea of paying for quotes seriously, or even think if could happen here in the States but not in Europe. However, a comment about Frost is on the money: they told us that we had won an award, so we said 'great'. They replied, if you want to tell people who have won, then you need to pay us for the package -- which included a trophy, a photo opp and a press release. We said, no keep the trophy: we'll just spread the news about the award ourselves. They told us that they would only anounce the award if we paid for it."

We'd love to hear about other experiences like this.

IDC misses forecast by 1900%

Interesting read on IDC un-prediction in today's El Reg:

Le Misancalculation: A one act Itanium tragedy by IDC (The Register 17/02/06)

Extracts:
"As the artist's rendering points out, IDC does not have a sparkling track record where Itanium sales prediction are concerned. In fact, its 2004 prediction was 1,900 per cent above the actual sales mark. Some analyst firms can survive on results like that, and some can't."
"IDC never explores these issues. Instead, it relies on flaky data and embarrassing conclusions to create a study that could outdo its 2004 effort - a feat we thought impossible."

WOW, this seems to be a hard one to get away from...

AR 101 series: Analysts and press quotes

David published Tuesday this most post after Tom Foremski of SiliconValleyWatcher wrote that vendors pay analysts for quotes:

• Analyst Insight: Do analysts charge for press release quotes?

Quotes for hire, interesting concept...

We obviously agree with David that this would be utterly un-ethical, but it's important to put the record straight:

We've never, ever, heard of anyone asking taking payment for quotes. We are very keen to hear if Tom can substantiate his allegations. We bet he can't and thus should retract.

PS (20/02/06): David in fact confused quotes and awards, which Frost & Sullivan charges for. Still not really ethical but not the same as charging for quotes.

This brings us to the subject of this AR 101: how can you leverage analysts in the media?

Analysts are, often rightly, viewed as thought leaders. Reporters value their insight, they are an invaluable ally to quickly nail down the issues around a topic and provide quotable material -all under tight deadlines. Journalists don't have the depth and experience of analysts, so a little help is welcome (an AR manager was heard saying recently that he believed the average VNU reporter age to be under 25...). So it's not a surprise that trade press often features analyst quotes.It's a good deal for analysts, they get exposure and it feeds their demanding ego. Some firms reward analysts for speaking to journalists, META was an example. This can lead to quote-happy analysts, some IDC analysts are particularly prolific as they produce so many cuts of the same data that every vendor can be a leader somewhere (read Give 'em all something - we need to sell reprints). Of course, too much goodness eventually hurts the analyst credibility...

So, how can AR managers leverage analysts to help their employer's profile in the media?

The most frequent tactics are:

• To include analysts quotes in press releases. This is subject to approval, Gartner and IDC have formal processes ; Gartner and AMR only allow quotes from published material (but not FirstTakes or Symposium presentations).


• To include analysts contact details in the press package sent to journalists, possibly with quotes.

PS (20/02/06): we received comments from AR managers and analysts saying those practices were highly ineffective -some journalists tend to deliberately ignore proposed quotes and go for "really independent" quotes. In other terms "shoveling quotes down a journo's throat will backfire and damage both vendor and analyst reputations." We would be interested in your comments on this...

Some additional tips:

• Depending on the relationship the AR manager has with them, independent analysts will often be quite willing to provide quotes and be contacted by reporters.


• It goes without saying that good AR managers will make sure that the analyst is briefed before reporters get the press release and that they have a (positive) opinion on the subject and contribute to the debate.


• Good AR managers will also ensure that they use analysts that are consistent between what they tell reporters and what they publish.


• Finally, some analysts are better at this game than others. A pedantic analyst speaking in 80 words sentences may not provide good quotable material to journalist and may even be mis-interpreted.

Other post in the AR 101 series:

• AR 101 series: Analyst vs. press (Analyst Insight)


• AR 101 series: the research process

Ovum to IPO, Forrester loses sales VP

Dave reports here that Ovum is about to float:
Analyst Insight: Ovum announces float

This was in the air for a long time but it's official now. We heard they stopped recruiting analysts/expanding because of this upcoming float, which is somewhat strange.
Anyway, the Ovumites are about to make some paper money as they all will get some shares -which is probably good as Ovum as a reputation for not paying them very well....

The other news today is that Forrester is loosing Tim Royston-Webb, its EMEA Sales VP (ex. Giga before), to Datamonitor. Both have a reputation of being pretty harmless at selling to vendors (check C'mon guys, sell to us!), so we think this change of ship is going to make no difference whatsoever.

The press release can be found in Tekrati:
Datamonitor Appoints Tim Royston-Webb as Managing Director, Technology Global

C'mon guys, can't you write-up a little the stories instead of just reprinting what you've read elsewhere?

KCG oppose, and offer, share of voice analysis

Our friends at KCG have us modestly confused. An article in their newsletter is titled Under the Influence: Five Reasons to Worry About "Share of Voice" Metrics.

Stephen explains that "we see most of our '“competition'” nowadays coming from PR firms. In attempting to transpose their typical volumetric 'Share of Voice'” measurement systems from the media business to the analyst influence business they regularly do our function a grave disservice."

In essence, he feels the key weaknesses of the share of voice approach are:

1. It is mistaken to measure share of voice inside firms; instead, track your top ten analysts.
2. They mistakenly measure the media: instead search research, which is essential.
3. Searching in the same way will produce different results at different firms; each firm needs to be searched differently to produce results that are consistent.
4. Searching on company name doesn't work; use key words.
5. Share of voice analysis does not only analyze quotes.

The criticism seems over-blown, because KCG actually sells share of voice analysis itself, what it calls 'Coverage Metrics', and so do most of its competitor PR agencies and the AR consultancies. These studies also measure volume, even if they eliminate the media and most analysts from consideration. The fact that KCG itself supplies share of voice studies is not the only clue that those metrics can be useful.

1. Sometimes it is appropriate to target firms and not only the key analysts. Much of the research about our firm is co-authored by a team of analysts. Following only one would be mistaken. It als would not fit the global spread of our business. We've learnt that analysts at the same firm but in other countries do not display the same preferences, and we need to get a top-level picture of the whole coverage by each firm, not just one who knows us best and is normally the most favorable. Even KCG stresses that its metrics "Measure coverage by target firm or target analyst list."
   
2. To not track analysts in the media would be mad. Analysts have a huge impact on the media and their influence is reflected differently in the media from in their research. To ignore the media is lazy and woud get us killed internally. When PR and sales come back with negative media clippings, how can we say they should not be tracked?
3. Searching is not that hard. I can recommend cut and paste: when I have the search right at a firm, I paste the search term into the back of the Excel sheet that our intern uses when she does the searches. It's very easy to be consistent if you think about it.
4. Searching on company name can work if you and your key competitors are in the same industries. Yes, if you're in many businesses you need one or more key words to reduce the number of results. But that's not hard. Use the cut and paste tip I mentioned above.
5. Quotes and share of voice are different things, but doesn't mean that only one of them can be useful. If my company is mentioned more in analyst research, or by analysts in the media, that's normally a better thing than being mentioned otherwise. Of course, it's also important to track quotes (although that means looking at the media, and breaking one of the KCG's commandments) but we do that by hand in a separate analysis.

In short, share of voice metrics are very useful and, honestly, if they were not then KCG and its competitors would not sell them. I am sure that KCG has a point that some PR agencies get it wrong at times: but if KCG is telling its clients to not track the media, then they will also be getting it wrong in the eyes of some clients, whose customers follow the media.

More reactions to the Information Week Credibility of Analysts article

Interesting take on InformationWeek's Credibility of analysts really bland article from Mike Rothman on his pseudo-blog (he does not allow comments):

• Educated end users need to know the context of a report (is it sponsored or not?) and treat the information accordingly.
  
• Having an analyst participate in vendor web casts, etc. is doing a service to the user community. Any analyst says a vendor's stuff is great on a own web cast is an idiot.
  
• There was no substantiation of the "small vendors only get covered if they are clients" claim.
  

IW had a great opportunity to educate their readers on how and when analysts can add value. Instead, they focused on old rumors and innuendo of bias. IW, if you are going to go on a witch hunt, you better find some witches. They are out there, and exposing clear instances of bias would have been interesting. They should have used the space to show clear examples of bias. That would have been interesting.

Duncan also commented on the same subject: ANALYST EQUITY: Research suggests Karma beats conspiracy:

• In outline form, Aberdeen's story is the same as in every Greek tragedy: pride; error; downfall; insight. Balancing between users and vendors is a difficult and dangerous job: few succeed for obvious reasons. As the Italians say, "He who serves two masters must lie to one of them."
• So we think that InformationWeek is right to point out the tensions; and we think that most analyst firms that try to serve both buyers and sellers do so at their peril. However, we also think that buyers and vendors both wise up to that reality quickly.

The Credibility Of Analysts [InformationWeek "blog"]

Larry Greenemeier from InformationWeek is running an interesting poll on the Credibility Of Analysts.

He plans to publish the results on the 6th of February.

Although is blog is not a real blog, it raises interesting issues, like why does the Borg not put products they place in its Gartner Magic Quadrant to the test.

Selected quotes:

"an analyst industry that's tapped into an almost desperate need that businesses have to stay on top of emerging technology and a community of IT vendors even more desperate to make the sale"

"IT vendors and their customers both became beholden to the big analyst firms"

"The reluctance that vendors and end users have in publicly questioning analyst firms tells us all we need to know about the clout that these firms have"