Vineyard Business Intelligence

My Business Intelligence links

Admin — 2010-03-06T14:51:54+01:00

As I promised quite a long time ago, I make a list of the Business Intelligence sites I regularly read.
I'm not always on the topics being discussed, but I learnt a lot for sure from them. I'll never thank enough those people who share their knowledge online. This is by no mean a complete or a popularity based list, and is grounded only in my personal taste.

BeyeNETWORK is the battleship of BI sites. Many of the most famous BI stars write here or their blog is hosted here or is aggregated here. Bill Inmon, Jill Dyche, Claudia Imhof and many others. Articles cover BI at 360 from high level philosophy discussion to low level implementation, from market news to change management in BI projects.
Being a rather cynical guy, the one defect I find that it is a little biased toward large, enterprise-size analysis.

TDWI is just an inch back from the previous site. It is not a totally informational site as it provides education and consulting, and it is better enjoyed with membership.

If I could, I'd marry this girl: Inside The Biz by Jill Dyche is one of the most insightful sites I have ever came across. She deals with a variety of topics, not all BI related but each and every article is worth reading.

For those who can read French, Le Grand BI is an immense enjoyment. Their prominent feature is a cynical and disillusioned approach to all the hype and buzzwords that sometimes blind us. I often refer to them to have a second opinion.

As I work for a large part in Microsoft ecosystem I use and like the Microsoft Resource Center . What's outstanding here is that you are taken by hand and brought to the highest level seamlessly by a host of written and filmed footage. Donald Farmer articles are a must read for all those who work with MS products.

Last but not least Oz-Analytics provides an always interesting point of view. Starting from other articles or real life experience, I've never read a trivial point of view from this guy.

As you can easily notice I leave out all those bloggers who deal mainly with the technical side of the job. Actually there are a lot of them worth reading, some fundamentals. This a good topic for another post, I'd say.

Enjoy.

In defense of the Natural Keys

Admin — 2010-01-21T11:46:42+01:00

Whenever I use the term "Natural key" in a tweet or on Facebook, someone always DM me to remind me how natural keys are worthless. I should use the brave and muscular surrogate keys for each and every task. Maybe is a frame of mind, but whenever there are strong technical debates I feel naturally inclined to the minority.
In this case my point is that natural keys are perfectly adequate for many Business Intelligence tasks, and should be replaced by surrogate keys only when necessary.

If you are still with me, let's go.

A bar counter definition of natural key is: "the code which uniquely identifies your record". Every transactional system assigns codes to its entities, customers, salesmen, products etc. The policy used to assign this code, is up to the transactional system and I could not care less.

Given that natural keys can a) change over time b) have a business meaning that can change over time as well c) not be the ideal field format for a join, often a surrogate key is introduced. A surrogate key is a key, usually numeric, unrelated with the natural, that uniquely identifies the record. That is, it does exactly the job already done by the natural key. (if you want a less mundane description of this subject you can find a good article here).

This consideration alone should drive to the conclusion that the widespread use of surrogate keys in a datawarehouse is often overkill.

I made literally tenths of systems with no surrogate keys which worked fine for years. Often we think to DWs as multi terabytes systems which integrate data from hundredths of systems, but these are only a fraction of the total. There's also a midsized business market which needs the bulk of the abilities provided by a datawarehouse with far less data and far less systems to be integrated. In these cases, a simple approach, based on natural keys, may be perfectly suited.

Building surrogate keys, is often a rather complex task, which slows development, hinders the long and delicate subsequent trimming work and slows loadings. More subtly, they're an entirely technical subject, hard to be understood by non technical project sponsors thus hard to justify on a project schedule.

Surrogate Keys should be introduced when they are necessary and there are, indeed, cases where surrogate keys are highly recommended.

In those multi terabytes DWs, where the smallest fact table has half billion records, performance becomes such an issue that the use of small numeric keys is practically mandatory. If many systems are integrated together in the DW, chances are that codes overlap, so an unrelated key is a good solution to that. Surrogate keys also shield DWs from changes of the naturals, whose propagation in a very large DW is an awful job. Also Surrogate Keys are a good artifact to factor in slowly changing dimensions. Etc. Etc. etc.

So, the next time I tweet the term "Natural Key", please do not flood my account with "Surrogate Keys"!

Quality content blogs: Startup and Software Entrepreneurship

Admin — 2009-12-21T00:25:16+01:00

Like many others, I regularly read a number of blogs. Reading, for me, is a way to relax after a tough mental effort like coding or a stressful consulting day. I even can't sleep if I do not read something. The written word is the only way to take my mind on a safe path before sleeping. No use to say I want to learn something from reading. Blogs are the main information source on the Internet. In my opinion the blog format is perfectly suited to the Internet use; it must not be as coherent as a book, it is shorter and it may have a touch of personal diary which closes the gap between the author and the reader.

I want to start with Startup and Software Entrepreneurship

Too easy to tell, but this is the place where I learnt that founding a successful MicroISV was possible is Joel On Software http://www.joelonsoftware.com/

Joel Spolsky has been the forerunner of software entrepreneurship blogging. He told the story of his company on-line, giving out some top quality content along the road. Some posts on software development are now classics and should be part of every developer's background. On the site some interesting forums are accessible; their vitality has decreased since when they require registration but a lot of high quality feedback can be accessed by posting there.

The other blog where every post is worth reading twice is A Smart Bear, from Jason Cohen http://blog.asmartbear.com/

Jason, a former software entrepreneur, has the incredible ability to always find something not trivial to say. I've never read something coming from him that wasn't absolutely insightful or eye-opener. A must read for each and every small software entrepreneur.

Penelope Trunk is a compulsive blogger who writes about herself on The Brazen Careerist http://blog.penelopetrunk.com/

She has the awesome ability to draw general conclusions and useful lessons from her personal history. Someway, she manages to be in the middle of something that has a meaning for you. She changes job, and you learn what you should do to get a new job. She's interviewed on TV, and you learn how to manage press releases etc. My wife reads her blog, just to learn about her life; I read her blog to learn about business life. Amazing! Sometimes we even talk about her at dinner!

My friend Peldi Guillizzoni runs his company blog, Balsamiq, telling the story of the business as it progresses http://www.balsamiq.com/blog/

Peldi is amazing, he was able to create a small multinational company (with just 2 employees!) out from his application Mockups. He often blogs about releases and bug fixes but, when he talks about entrepreneurship, he really deserves attention. I learnt from him how the Internet works today.

The last outstanding blogger I want to mention in this post is Eric Sink http://www.ericsink.com/

He's a software company founder with a very clear vision of what the business of software should be. Lately he blogged a lot about source control systems, but older posts on software product marketing are true classics. If you are a bit geeky or you come from a technical area, Eric's blog will effortlessly introduce you to those unknown realms.

In the next post, I'll talk about the Business Intelligence blogs that I regularly follow.

Feature Spotlight: assisted input and auto coding

Admin — 2009-12-13T13:41:20+01:00

This is the fourth and last post of a series dedicated to Viney@rd features. Find the first here, the second here and the third here.

One of the Viney@rd groundbreaking features is enabling the user to fully control the model content by simply compiling a worksheet in Excel. Two things, nonetheless, may be particularly annoying while doing this.

Viney@rd uses codes to identify the dimension key. If you compile the table by hand, remembering the right code for a dimension element may be a pain. So the "Assisted Input" feature is implemented. When turned on, a textbox and a list appear on the screen. Simply select the code column, type the few first letter of the dimension element description and it will be selected on the list, along with other matching elements. Simply hit enter or click on the element to add it to the table.

Rather than adding rows manually, it is likely the user will have to copy and paste data received from other sources. Often, human-made tables do not have codes but simply have descriptions. Getting the right codes is a very time consuming process. So, simply paste descriptions in the proper column and then click "Add Codes". The "Add Codes" command infers the code from the description. It works perfectly when the description is exactly the Viney@rd description, but can also "Guess" what the right code is. False positives are rare because, when it is "unsure", it rather leaves the code blank. The assisted input can help the user compile the codes left blank.

Now you can easyly test Viney@rd for free downloading it from here. The trial period lasts 90 days and it is absolutely free of charge and with no obligations whatsoever!

If you want to learn more about Viney@rd visit the web site!

Feature Spotlight: Automatic connections to Viney@rd database

Admin — 2009-12-05T23:22:43+01:00

This is the third post of a series dedicated to Viney@rd features. Find the first here and the second here.

Viney@rd has its own query feature, which can draw tables on Excel worksheets, but it is aimed mainly at creating complex, refreshable, layouts. It is easy to understand that, while this is felt by many as a requirement not fulfilled by classical BI tools, it is not the best approach to data analysis. Controllers and data analysts require both. Nonetheless Excel has a powerful built-in feature to analyze data, the pivot tables.
One of the ideas underlying Viney@rd is not to cater twice what the user has already purchased, so, instead of duplicating the feature, I created something to integrate pivot tables with Viney@rd.

From Viney@rd client you can create a workbook connection to a model. The connection exposes as fields all the dimensions, the attributes and the measures. So, you can analyze all these data in the usual way.

With this feature you can also use Viney@rd data as a source for PowerPivot engine in the coming 2010 version, working with an even better analysis tool.

Try Viney@rd, 3 months, fully functional demo!

Learn more about Viney@rd

Feature Spotlight: Automatic number detection

Admin — 2009-12-01T11:12:29+01:00

This is the second post of a series dedicated to Viney@rd features. Find the first here.

Viney@rd is not a mere reporting tool. While the ability to feed each single cell in a worksheet is very important, all the model building process made through MasterBook.xls has a very high value added. The key point is that no database knowledge is required and data can be saved within models directly by Excel.

To this respect, the user must not care too much about codes, Viney@rd can "ingest and metabolize" almost everything. In particular, you can have a dimension with both numerical and alphanumeric codes.

Albeit you can save everything into the repository, the nature of what you save influences the outcome of the filters that you place on the query. The filters with" <"and" >" signs may not work as expected. If you have mixed alpha and numbers like 001, 002,003, ToBeAssigned, InAssignement etc, there is no choice other than comparing alphabetically. If you compare alphabetically, the system does not know that numbers are numbers, so they will come before letters and 10 will be before 9. The best way to avoid this is padding the numbers with 0s, that is 001 instead of 1, 013 instead of 13 etc., a rather tedious work.

Nonetheless, Viney@rd is smart enough to recognize dimensions, or attributes, which are entirely numeric, and compare them accordingly.

So, the user must not worry about things like defining objects as numeric or alpha, or padding the numbers to sort or filter them correctly .

Feature spotlight: query synchronization

Admin — 2009-11-25T13:51:09+01:00

This is the first post of a series focused on a single Viney@rd feature, to let you know about those features which are not immediately visible but, all together, build up Viney@rd strength.

�

Let�s start with query synchronization.

�

In a workbook you can often have two similar queries, which differ only from few particulars, mainly conditions. A typical case is the current year vs previous year or budget comparison.
A less trivial case occurs when comparing not homogeneous data. For example, you have main query with customer and sales, but your customers are both other companies and individuals. For companies you have a credit model showing the customer�s financial situation, for the individuals you have a model about customer�s demographic. You want to combine the three models in the same spreadsheet area.

�

The synchronization feature let you sort and filter the results of a query according to another query.

�

Please note that this feature is hard to find even in very expensive reporting tool and is never implemented in such a pragmatic manner.

�

For example, you may wish to get your customers YTD value and have the corresponding budget aside. Invoices and budget will likely be in two different models, and the customer�s order is not guaranteed to be the same; more, new customers may be present in the YTD query but not in the budget query. Syncing the budget query with the YTD query means that Viney@rd will scan the first location of the YTD query and will output the budget data in the same order as in the YTD query.

�

YTD Query
Customer	Value
�Customer A	7000
Customer B	6500
Customer C	3000
Customer D	2000

�

Budget Query
Customer	Budget
Customer C	3500
Customer A	5000
Customer B	5500
Customer E	500

�

If the budget query is synced to the YTD query the result as printed on Excel will be

�

Combined Result
Customer	Value	Customer	Budget
Customer A	7000	Customer A	5000
Customer B	6500	Customer B	5500
Customer C	3000	Customer C	3500
Customer D	2000	�	�

�

Notice that budget customers are now printed in the same order as YTD customers. Customer D has budget columns blank as there was no corresponding member in the budget query results. Customer E, which has no YTD value, is not shown, thus causing a filter effect.

�

Please notice that custom sorts may be implemented with this feature. For example: refresh the YTD query, sort the customers in the order you want, refresh the budget query. The order in which the budget results will be printed is the same as in Excel, but Viney@rd always start scanning from the first cell if the position of the YTD query as defined in Viney@rd.

�

Notice also that we have now two customer columns. This layout is usually redundant and unwanted. To hide the second query dimensions, uncheck the �Show Row Hdg.� Checkbox. This checkbox also works with stand alone queries, preventing the row headers to be written on the worksheet.

�

Synchronizing two queries is an easy task, just create the two queries, go to the query to be synced and pick the main query in the drop down. That�s all!

Four steps to positioning your software within the market

Admin — 2009-11-22T00:13:38+01:00

Some software entrepreneurs, when asked, say that their market is whoever buys the box. Some have a clue about who their customers are, but they really do not know well. Some others have a razor sharp knowledge of their market and carefully drive their application development: they know their market positioning. Market positioning is not something that happens, it is a choice that should drive your development efforts. So, it is worth knowing something about. In this post I will even resist the urge to explain where Viney@rd is positioned within the market, so keep reading safely.

First, what is market positioning? The best way to understand it is to think to the market as the shelves of a library. Each shelf covers a topic or an author. The books in the front rows are more accessible than those in the back rows. The books on the higher shelves are more difficult to reach than those at eye level. People looking for a book will locate the appropriate shelf, then they'll likely grab a book on the front row. If the shelf is located 3 meters high, chances are they'll renounce getting it, unless they exactly need that book or a book from the top shelf.

Some people will pass by and casually spot an interesting title and they'll get it. Some others will require a specific book and will browse through the entire library to find it.

The key point, anyway, is that a book (that is, a piece of software) can stay on one shelf. There's nothing like a "matrix oriented shelf".

So, let's start from the last point. Some of you will say that some package is good at doing two or more things. Penthao, for example, contains ETL, reporting, ad hoc query etc. Open Office is good at doing word processing, spreadsheets, database etc. Wrong, Penthao lies among the Business Intelligence suites with Business Objects, Microstrategy, Cognos etc etc; the one thing it does is organizing the process of consuming business information. Open Office is an office automation suite, it organizes the way an information worker relates with knowledge and fellow workers. Even if, technically, a piece of software does more than one task, it will be perceived by the market to fit into a single category. This happens because people define things by comparing them, so anything is something else with some differences (my wife is used to say that a rabbit is cat with long ears and a short tail, but my wife is graduated in social sciences, not biology �).

The library of software, anyway, is organized in two large rooms, the consumer and the business markets. These are very very broad distinctions. Some find difficult to draw a clear line among the two but I use a definition that appear to be sensible, at least to me. A consumer software addresses the problem of an individual or her family, a business software addresses the problem of a business entity, no matter how small it is. The consumer market has looser requirements (if a video game crashes, nobody is hurt and no money is lost �) with low unit costs, usually low unit margins and high volume. The business market is more varied, ranges from few bucks invoicing softwares for professionals to multimillion, high end, enterprise management systems. Often it has higher margins and must comply to much more compelling standards.

First choice: will you build your software for consumer or for business market?

From now on I will cover only the business money because a) it is the one I really know b) it is where the real money is. In my opinion it is the place where even a small or a micro company can prosper because it requires volumes much lower than those of the consumer, Userscape, SmartBear, Balsamiqor FogCreek are living examples of this.

So you enter the business room and you will soon realize that the vast majority of the books is in English. There are few shelves of books in foreign languages and few books which are present in the library in more than one language. It's no use to say that the vast majority of the readers picks books in English. Translation: Microsoft Windows rules. All the other OSes have their role in the market but they occupy a niche. This is not going to change in the near future. I know that I'll be flamed for this, but it's a fact of life. So, the operating system is not really a choice, while being multi platform is a plus, not building for Windows imposes heavy self limitations.

The actual choice though, is related to what is called "ecosystem", that is, the complex of business software and architectural building blocks that revolve around one of the big players. The ecosystems worth considering are:
Microsoft (AX/Nav - SharePoint - Office - SQL Server - .NET)
Oracle (Oracle Apps/One world - The Database - Hyperion - chunks of Linux)
SAP (The ERP - Netweaver - Business Objects)
There are also minor ecosystems which feature a less comprehensive coverage of the business world like:
IBM (websphere - db2/Informix - iSeries and X worlds)
Open Source (a blob of technologies revolving around Java and the LAMP stack) and few others.
To be complete, I must say that there are technologies which span through all of these ecosystems (like Adobe, the MS Office application or the rising Google) but are rather specific and not really relevant at the moment of this writing, despite all the hype which surrounds them.

The ecosystems imply that, in a business environment, an ecosystem tend to dominate the others; that is, the "best of breed" approach is always less and less popular because it requires interfaces among systems, which imply costs and complexity.

The ecosystems are not perfectly equivalent despite the marketing claims, so you have to pick carefully because your choice will focus you toward the referring market. Even the most general applications have a bias toward it. For example, if you build a LAMP stack based application for meeting rooms management, you may feel to be completely independent. In a business environment, anyway, you'll be asked first to interface your data with the local ERP HR module to have evidence on the employees data of the meeting usage, then with MS Outlook to coordinate reservations with meeting invitations, then with the accounting module to automatically post meeting costs etc. At the end, the market will ask you to have your PHP created screens inside the ERP, and you'll find yourself locked within an ecosystem, even if you did not mean to. The more the app you are creating is complex and the less it can be easily ported to other environments, the more you'll be compelled to chose a reference ecosystem.

Second Choice: which ecosystem do you pick?

So far you have chosen a shelves row in the library, now you have to find a position to place your book. Of course, the books on the front rows are spotted much easily and chances are that they will be the most read. Nobody picks that dusty gray book in the back row, unless she needs exactly that one.

That is, your software may address a wide range of requirements, standing together the other players on the market, or it can focus on a specific niche with specific requirements, becoming known only to niche players. From a business perspective, both solutions are viable and can lead to a sustainable cruise speed. What you have to do, anyway, is rather different. In the first case you'll need to tie, more or less, the other players in most of the features and to be outstanding in the feature which differentiates you. In the second case you need to concentrate on your niche and to adhere closely to your (potential) customers requests. The mindset, the competency, even the people you need are different.

Third Choice: do your software addresses to a niche audience or to the broader public?

The books on the higher shelves are, of course out of reach of many, none the less few people still read them. Even if your software is addressed to a niche audience, chances are that there are going to be different price ranges. You have to chose the tier in which your software will lay. Obviously, the low market tier will target lower unitary margins and higher volumes, the hi-end market will feature few sales to few key customers. As explained before, you need to work differently to address the two segments. Be aware that there is a point not obvious about this. At lower ends you usually end up talking about features, at higher level what matters most are the concepts inherently embedded within the software; that is, how the business processes are influenced by the use of the software (no matter if there is auto completion for fields or not).

Fourth Choice: does your software addresses the low, the mid or the high tier of the market?

So, simply writing down the four questions and answering them clearly in one or two sentences may well help you decide where to go with your software.

Did you do this kind of analysis before starting up your company? Do you have any experience on this kind of planning? Let me know.

Viney@rd and PowerPivot: no overlap but synergy

Admin — 2009-11-13T12:48:05+01:00

I've been asked about how Viney@rd relates with the environment that Microsoft is designing with PowerPivot, Excel 2010 and SharePoint 2010 upgrades (Gregory and Rob, thank you for considering!). I didn't have had the occasion to dig deep in all those applications yet, so I want to lay only few points that appear quite acquired so far.

I must confess that there has been a moment when I feared that Viney@rd was doomed to become a poor copy. Luckily, some of the ideas which Viney@rd is based upon are such that the others are too snob to implement them ;o)

1) Power Pivot takes Excel ability to handle and analyze the result of a complex dataset to the next level. Viney@rd, though it can organize query results in tables, can address a single cell or a single row, removing headers from tables. So complex Excel layouts can be preserved upon query refresh with no intermediate "database" sheets or VBA.
This reconciles Excel layout versatility with the ability to refresh data from the DB.
I've seen so many controllers or sales analysts flaunting an Excel printout at me saying: "I do not need drill down, I need this. It takes half a day to make, make it refresh with the click of a button", that I decided to build the app.

2) PowerPivot can coalesce data from multiple data sources in a single, high performance "blob"; Viney@rd gently takes you to incrementally creating a datamart federation. The business user, likely, is not even aware of what's doing. PowerPivot requires a database working knowledge, Viney@rd does not require any.
Anyway, nothing forbids the experienced user to consume Viney@rd data from PowerPivot. She can go on tables directly or use the flat view which is created together with every model.

I'm also looking to PowerPivot high speed engine as a mean to boost Viney@rd performances.

3) The entire Microsoft's castle is devoted to reporting, ad hoc query, data analysis etc. but cannot modify data. With Viny@rd you can move, change and enrich data with no technical knowledge.

All these points deserve more attention but, by now, Is unwise to go deeper without a thorough analysis of what the new applications can do. I'm just waiting to see some nearly final versions to run an extensive test. Meanwhile Office 2010 is not supported,

Enjoy

I Always Find Something

Admin — 2009-11-11T01:08:32+01:00

When I'm working on Viney@rd I spend quite a large amount of time surfing the web, checking blogs on Twitter and Facebook. I try to connect with people who could be interested to Viney@rd, of course. Seen from the outside this activity appear to be a waste of time.

My wife once observed me and drew this conclusion. Actually, once, I agreed with her. It was a way to pause in the intervals between tasks. I theorize that people have a fixed efficiency, and there is a limited amount of time to be spent "in the zone", before giving up to something not requiring the same concentration level.

Working in this way, I slowly realized that I started finding some small nuggets here and there. It might be photos, a news fragment or a blog post, all of them worthy somehow. The Internet is a powerful knowledge aggregator and exploring the topic suggested by the accidental finding is often very very easy.
So I learnt of things I didn't know about, like truffles restaurants in New York or Win 7 vulnerabilities. This is called Serendipity, but this term often refers to science discoveries. My serendipity is small an intimate, and makes me a little better, one bit a time.

If you feel like yelling at me because I do not talk about Spreadsheet Automation and BI, do not do it here; download and setup Viney@rd and use the "Complain" feature :o)!

Take Care

Version 1.1 is here

Admin — 2009-11-06T15:41:12+01:00

Building version 1.1 has been a tough work, that's why I disappeared for a while. New features have been added but, most of all, a lot of testing has been done, making this release a lot more stable than the previous.
The leit-motiv of this release is "Listen to the users".

�

First, I focused on making the software easier to configure.
Working with SQL server is at the root of the issue: non technical users must always find out a SQL instance to connect to or install a free version on their machine. Both these tasks are not trivial for some business user.
Unluckily, there are few things to do about that without radically changing the product proposition.
I've been suggested to support MS Access as well; I do not reject the idea in principle but it requires a fairly large amount of work and automatically downsizes the application.
So, be prepared to see some news on this topic in the near future.

�

In the meanwhile, I introduced a configuration wizard that guides users through the configuration steps in a colloquial way.

�

Windows authentication is now supported, making configuration somewhat easier, especially after installing the SQL server express versions. From a technical perspective, anyway, lacking this support was a big turnoff.

�

There is a totally new set of demo data, more consistent and sensible, based on Microsoft's Adventure Works database. Some users may also be more familiar with it.

�

The new "web bar" feature is aimed at communicating with the users; given the downloads I had, I got too little feedback. So I encourage all of you to make use of the new "Complain" feature that will land directly on me everything you need to tell. Actually I was thinking to name the feature "Yell at me", but it looked rather unprofessional.
Beside the complaints, you can access directly the knowledge base, the user forum, this blog and the user manual. Do not hesitate to use the forums either, too few use it now and mail me directly. I think that sharing your thoughts should be an advantage for everyone. (actually, I've been told that the forum look is too old style, but I'm working on it, just be a little patient!).

�

On the client side, I added an option to hide the query row headers. This is useful for synchronized queries placed side by side, like previous versus current year. I will never stress enough that this feature is very hard to be found in other query packages.

�

Finally, a slightly better look has been designed and a number of bugs have been fixed. See the knowledge base to learn about the fixed known issues.

�

So, I release this version to you. I'd really love to hear your feedback. If you buy Viney@rd now, I'm happy, if you don't and you want to explain me why, I'm even happier because this gives me a chance to make you change your mind.

�

Enjoy

Why you ALWAYS need a specific database for business data analysis

Admin — 2009-10-15T21:47:44+01:00

It is often said and widely acknowledged that a datawarehouse for business data analysis is something for the big guys. I do not agree at all! Having little data does not mean that the environment is not complex. I had small customers which required very advanced analytics to keep up with what is going on.
So, I try to explain as simply as I can why you need a database for business data analysis (i.e. a datawarehouse) EVEN IF you are a small or medium company, have few data and your server can handle all the workload etc.

1) Your history is there even if your OLTP systems blows up. It's not a matter of backups: the vendor goes belly up and there's no further support, your business changes too much to keep using the old system, your new CEO loves a different SW etc.

2) Building a datawarehouse forces you to think to an analysis model for your business. A stable model makes comparisons with the past possible. The need for changes in a data analysis database mirrors the need to change the company strategy; changing too often means that your company does not know where she's going.

3) Building such a database forces you to think to key performance indicators for your business. There are figures for every business; "magic numbers" that give you an immediate idea about what's going on. No business, no matter how small it is, can do without some key performance indicators. Often, they're not formally defined, but they do exist in people's minds.

4) Likely you have different systems with different master data. The datawarehouse is the place where you can match all your siloed data . All those small MS Access databases, MS Excel worksheets etc. that you have around generate an inextricable mess to understand what's going on. If you build such an analysis database makes you actually tackle all the different data and reconcile them.

5) Often business people think by categories not implemented in a business application. Your datwarehouse may be the only place where some data may reside thus being applicable to your analysis model.

6) having a different DB, maybe on a different technology, make the analysts feel special compared to all those data entry people, and make it more acceptable to the upper management.

As you see, the point is hardly technical but is related to the way a business looks onto itself. A datawarehouse , to such extent, is like a mirror in front of which you business can realize that there's that little spot on the neck.

Do you have examples about datawarehousing in SMBs? Do you share my vision? If yes, can you define it better? If no, why?

The BI vendor that SUCKS: part 2

Admin — 2009-10-10T14:39:03+01:00

In a previous post, I already talked about a BI vendor I came across in my "other" job as a BI consultant. Maybe one day I'll disclose the name but now I can't, thus I'm using the fake name "Sally".

Their product has strength and weaknesses, but overall is a good choice, in my opinion, for managerial accounting. Actually, the decision to support Sally has been taken at a level higher than mine so I have to comply.
To consolidate this new partnership, I spent a couple of days with Sally people, got a bit of training and invested few other days experimenting, in the view of a project we'll start in the coming days.

I focused on the technical side but, en passant, I was sitting in the same room of a briefing with the sales and marketing people.

Then I saw The Slide.

The title said "Commercial strengths and guidelines".
The points were:

Do not talk to the IT, they'll easily spot our weaknesses (namely, we are non standard in many areas)
Talk with other managers, they'll appreciate our beautiful clickable charts
Tell the managers it is so simple to manage that it does not require IT involvement
With Sally you get instant customer lock-in, very few know it thoroughly so switching suppliers is very hard.
It is complex enough to always require a consultant intervention for every small change, so you can sell more man -hours.

I'm old enough to know that these kind of things happen in the jungle out there, but I keep thinking that this attitude is the root of the evil. What astonished me was such an open display of ethically questionable points and many people in the room actually appreciated them!
Later, I cautiously probed one of their marketing people and she was not concerned at all.

In my opinion, this attitude should be totally rejected because:

It is ethically questionable; I'm here to solve problems and provide value to customers, not to "grab them by testicles", as a salesman said.
It is terribly short sighted. Drawbacks will emerge, someday, and long term value from the customer will be lost.
It expresses a supreme contempt for the customers, who are treated like kids.
It turns what actually are weak points to "artificial" strengths.

At this point I should draw a final conclusion. I do not have any, Please express your opinion and share with us some similar experiences.

Enjoy

How a startup can change your life for better

Admin — 2009-10-01T12:42:24+01:00

Since when I started working on StraySoft, my life has changed. Should I stop this adventure now because alien abducted me to Mars, I'd bring in space with me a lot of new experiences.

First, I do not watch TV any more. TV is an incredible time waster, I realized I used to turn it on only because it relieved me from thinking. We believe that, sometimes, we need to stop thinking and let our brain rest. I think this is a fake. As long as you think to something you're passionate about, you do not need any mental rest. TV is a poor surrogate for using your brain in a pleasant and productive way.
Actually I pick carefully what I want to see (mainly journalism and sports) and I skip the so called "entertainment" flowing out of the screen. Far from being tired, my brain has never been in a better shape.

I learnt to use my time much more efficiently. StraySoft requires from one to four extra work hours a day, and 10/12 hours on weekends, hence I have to reserve time for my wife and other "routine" tasks. Since when I begun with StraySoft, the quality of the time I spend with my wife has increased substantially. We no longer sleep together on the couch in front of the TV set but we actually talk, or we do something, or we go somewhere together. She is not completely comfortable with the situation but yet very supportive, and she does not dislike the "fast track".

I learnt a lot on entrepreneurship. I've always been a salaried worker but, being a consultant I have rather deep knowledge of the process that make a company work. Nonetheless I realized that entrepreneurship is different. The glasses you wear are of a different color, and the angles you discover are otherwise hidden. It's much more catching. Much more fun.
Along the way I met a lot of people who enriched me in a way I thought it could never be possible. They are so different from the people I was used to hang around, their common background is "it can be done!" instead of "is there a way to f*** the others?".

I keep my consulting job totally separate from this effort, but my performance dramatically improved; I'm more focused, fast learning and assertive. I hope I won't get promoted because I have no time to steal from StraySoft!

A new interview

Admin — 2009-09-26T21:15:58+01:00

It appears that someone is starting to think that I'm a rockstar or sort of :o)

Ajay Ohri from DecisionStats interviewed me about the company and BI in general.
Thank you Ajay for giving me voice!

Please find the interview here

Interview by Mike Johnson

Admin — 2009-09-24T23:23:27+01:00

I've been interviewed on BI and Viney@rd by Michael Johnson, who run a really interesting blog.
Thank you Mike for giving me voice!
Please find the interview here

Visit Mike's site at http://www.bigattichouse.com/

How a BI vendor can instantly SUCK!

Admin — 2009-09-18T23:23:53+01:00

I just came across a BI app vendor that shall remain nameless. In the last weeks I had the occasion to evaluate in depth their product. Like many others it has strengths and weaknesses and a few outstanding points.
Among the others, it features a scripting language that can be used to make custom drawings, graphs and forms. The setback is that the language has no commonalities with any other scripting language and is rather obscure. Anyway, for the project I was working on, overall, this product could be a good choice.

Browsing documentation I noticed that there was no reference to the scripting commands used to draw all those beautifully looking forms, with nicely rounded buttons and combos to make selections and spawn graphs everywhere. So I contacted the authors to ask for documentation. With my total surprise, I discovered that the vendor does not disclose those commands and their documentation to third parties. I had quite a long argument with one of the managers, till he reluctantly admitted that, offering themselves professional services and consultancy, they wanted to maintain a "competitive advantage" over their external developers and partners

In my opinion this is unethical and plainly stupid! It is hardly mentioned but the first element in a vendor-partner or vendor-user relationship is trust.
I must trust the vendor that every reasonable effort to create a stable and usable product is made.
I must trust the vendor that whatever issue should arise, any reasonable effort to solve it is made.
I must trust the vendor that there's always going to be a smooth and voluntary transition to version +1 (a world famous player broke this rule few years ago, risking a lot!).
I must trust the vendor I'll be told about any feature and technicality I need to do my job and make successful projects.
If all these conditions are met, then I can evaluate the product and actually use it.

I'm going to comply to these rules as much as I can; it is a commitment I feel unavoidable. In the next few months I'll make the partner program available and partners will receive the maximum assistance possible for a small fee. I want that those who believe in me and in my product will be successful with their job. No more, no less. I will never refuse to share all the knowledge required to do a project.

Just for the records, the only thing I got from that vendor was a sample application to use as inspiration. Better than nothing.

What's your opinion? Do you have similar stories to share? Let me know! Your comments are welcome!

Version 1.0.1 released

Admin — 2009-09-14T16:14:45+01:00

Welcome! From today (09/14/2009) Viney@rd version 1.0.1 is available!

What�s new in version 1.0.1?

The aggregation transformation has been re-implemented. Now the user has more control over the transformation, both targeting existing models ore creating new aggregation.
Old transformations are still valid but require editing before being run again!

In previous versions, for models created by copy or aggregation, the data connection for pivot tables creation failed. This bug is resolved.

A fast visual effect has been introduced in tabbed forms, to improve the overall look and feel.

Other minor bugs have been corrected.

Go here to download the new version. Simply install this version over the old one to update the installation. Refer to the user manual for further details.

So far the institutional communication, but I would like to say a few words about releasing a minor version.

It appears that a minor version should be an easy release. Version 1.0.1, actually, corrects a few bugs and streamlines a feature that was already implemented.
Completing development has been rather easy. Completing the other tasks was not as easy as you could think.
StraySoft is an extremely lean company, so I did everything myself, and yet it took a full day to cope with all the details.

First I had to update the �About� screens on both the documents.

Than I had to update the manual. I changed the Aggregation Transformation chapter, than I added a �What�s New� section and an �Update Section�. I also updated the screenshot in the manual.

Than I had to rebuild the setup package, removing the old files and adding the new ones.

After creating the new packages I tested setup with XP and Vista, both on a clean machine and as an update.

At last I updated the web site to let you download the new versions and leave the old available.

You can easily understand that it is a fair amount of work. Sometimes it happens to read rants about �that simple application update that is clearly six lines of code, at can be done in five minutes, and if you do not do that you are a �. Maybe it�s five minutes coding, but weeks of other tasks. Think of it.

Enjoy

Making The 1.0 - part 3.0

Admin — 2009-09-06T22:29:46+01:00

In this third and final part of the road to making the 1.0 I�ll shortly describe all the other tasks I had to accomplish before releasing the application.

I decided to add a forum to the website so people who want to discuss the application or ask for support have a place to gather all together. That�s easier for sure than relying on e-mail to support them.
I found an interesting free software that fits well to my needs.

I also used it to prepare a knowledge base. In the beta I included a �known issues� document but, in this way, I can provide and always updated information. I confess that, when I saw the forums up and running, I felt like I was creating a real company. All this experience is absolutely exciting and I love almost every moment.

Another feature I implemented is a pdf brochure to be downloaded. It is something that can be saved somewhere on the PC and can stay there to remind people about Viney@rd. I used Microsoft Publisher to create it and it was fairly easy. I�m not totally satisfied, though, as it can be more incisive.

A thing that took forever to implement was the screenshot page. Selecting all the right screenshots, placing them on the page, describing them accurately, was a process that took a whole lot of time. Nonetheless I think that this feature, even replicated in the PowerPoint downloadable slideshow gives the best insight before downloading the trial.

The last thing I did was configuring the e-commerce platform. I�ve chosen the Plimus platform that has many advantages. First, they do not ask for a fixed fee but simply charge a commission on sales. Secondly their platform is highly customizable by a smart proprietary tagging system. Most of all, they act as resellers, so you do not buy Viney@rd from an Italian company but from a well known world distributor.

Getting to the point of having an application ready for sale has been quite hard. The amount of �blood, sweat and tears� is not negligible at all. Countless hours went in debug, documentation, research, testing etc. The sensation I had when the Buy page went live is hard to tell, and was a fair reward for what I�ve done and for the people who helped me.

A new era begins.

Making The 1.0 - part 2.0

Admin — 2009-09-03T22:05:58+01:00

In this second part of my journey through putting online a software business, I'll focus on the new application name, the damage I did changing it, and all the myriads of tasks that must be accomplished to go online with a decent website.

After receiving a few site and overall marketing reviews I'm not proud of, I decided to dump the old site and the old overall look to get something new. I sat on the couch on the back of my desk with a notes and a pen and run a one-man brainstorming session.

First, I needed a modern looking layout, something with a web 2.0 flavor yet looking serious and professional. Applications like Viney@rd are often perceived as extremely serious (I do find them fascinating, instead�) so I can't be too easy-going. Nor I can be seen as an old mummy producing a dull and sad corporate accounting application.
This required to find a layout I could use, cheap or, better, free. I'm not a web designer and I can't afford hiring one for a full out project.

I also decided to be as simple as possible. The old design was terribly cluttered. Communicating business intelligence is difficult enough without all those widgets to take care of. There is no way to avoid long text explanations for those who need in-depth information but few clear points are a must if you want to put the reader through the site.

As a consequence of the previous points, almost all the site copies had to be rewritten. Focus on the product, focus on the benefits, focus on the customers; nothing else is of relevance. Deleting my photo from every site page was the single most significant improvement, making it much more enjoyable.

To look like a real company, the site needed at last:

A through product section
A complete case section
A knowledge base
A user forum
A downloadable brochure
An easy trial download page
A complete and professional e-commerce section

At that point my wife entered the room, saw me staring at the ceiling while pondering, and started explaining me that I had to do something with my life instead of staring at the white�.

The first thing I had to do was to find a template. I connected and browsed at least 300 different layouts from at least a dozen of designers. I checked the layout on the designer's site, than I downloaded those looking more promising. I unzipped 43 of them in a folder, than I let them alone for a couple of days. This is crucial because what you like at first impression may look stale and annoying at second sight. In fact, two days later I shortlisted only 4 layouts. The winner was the actual site design by First Light Web Design, that has the great merit of looking modern and simple at the same time.

So I came to personalizing it. I wanted a cool logo and a cool heading. The existing one was nice enough, but I wanted something more impressive. Actually I do not know how I managed to change the color schema in PowerPoint to find the present colors, but when I saw it I liked it a lot. I played around a bit and it popped out. It conveys a sense of warmth and looks rather "Californian". What do you think about?

Changing the name from Vineyard to Viney@rd is an entirely different matter. The idea rose from noticing that, in the header, simply writing Vineyard in red produced a rather static layout. So I painted the "a" in blue, scratched my head, made it capital, scratched again, noticed the @ sign on the keyboard and replaced the "a" with it. Even in this case, when I saw it I loved it. It's exactly the kind of simple creativity that I like.

So I merrily decided to dump Vineyard entirely to switch to the new format everywhere! I fired up the code editor, and run a search and replace in order to change all the captions. No use to say that I ended up with registry keys with the "@" sign in the path, or database names and directory paths entirely spoiled. "@" bugs started to pop out from everywhere and it took a week to fix them all (some were deeply hidden in the foundation code and subtly menaced the stability of the entire castle). That's not enough because the old name was even in the site, in all the copies, in the manual, in the setup, in documents (even the legal ones), in the breakfast milk cup, under the pillow, down in the garden, on the back of my eyelids etc. When I was half done I thought that, maybe, that name was not so cool but now I'm really satisfied and I'm evaluating the same naming convention even for the next suite component (suspense suspense, I'm not disclosing it now�). Please let me know if you spot a regular Vineyard somewhere.

Writing the product and the case pages was a tough job but I immensely enjoyed it. It's not easy to have the occasion to talk about your beloved product over and over again. I wrote and scratched various prototypes, but the scheme 1 hyped marketing page + specific user pages seemed to me a good approach. Some will say that I write too much but a tool like Viney@rd is not purchased on instinctive decision. I do not want it to be purchased without a full understanding of the use to be done with it. So I preferred to be completely open and describe it fully. Probably I'll miss some sales but I prefer to have more educated customers.

I'm not done but this post is becoming too long. In this case, there is no drawback in making a third post about the 1.0.

Making The 1.0 - part 1.0

Admin — 2009-08-15T01:09:44+01:00

I was asked about my daily work, especially how I do things.
Apparently this kind of information is terribly interesting for those who are willing to make the leap and going micro-ISV by themselves.
I do not think to be a "best practice" model; but I can say that some procedures work for me.

All of the Viney@rd development is made within Excel. Viney@rd is 100% VBA (but for few SQL setup scripts), so there is no reason to use anything else. VSTO will enter the game in the near future, but first versions will be pure VBA for sure.

I do not use any specific version control tool. I have a development directory and many version directory. Testing is a continuous process, as after every change of any relevance, I build a complex report using all the features. MasterBook changes are tested with a set of heterogeneous data.
The build process is very simple so I make the build only to release a new version.

I do not use any specific bug tracking and project management tool; MS Office Ultimate has everything I need to coordinate myself and those who help me. It was really exciting, few months ago, seeing each task turn green as the features were completed. At that time I thought that the go-live was near, but I was an optimist!

The first web site was built by Microsoft Expression. Actually it was really good for site management but produced a bloated html. It subtly induced a "Microsoftish" site design that was definitely not cool.
I received a number of bad reviews about the site, and so I decided to dump it and change over. I evaluated at least 300 templates and came down with a template from First Light Web Design . The actual work is done by the Coffee Cup Free Html Editor . It is terribly "skinny" compared to MS Expression but it has the features I need and I feel rather comfortable with it

Many will be horrified to learn that I created all the graphics by PowerPoint, using the "save as image" feature, adjusting sizes and converting formats by the "Office Picture Manager". I'm not a graphic designer, so I looked for suggestions here and there and realized that all the icons and graphics I needed could be better designed by PowerPoint. The new prebuilt styles save a lot of time creating a professional looking graphics. I'm ready to be flamed by Photoshop purists :o).

The new site will go online with the 1.0. I'm not publishing a preview because few key points are still to be defined.

In the next post I'll talk about why I decided to change the application name, the damage I did with that decision and all the tasks that I'm completing to release not only an application but a professional looking website.

A Blurred Vision of Truth

Admin — 2009-08-04T12:33:32+01:00

Viney@rd will fit, at least by now, inside the broader "workgroup intelligence" category. This is often criticized because of the lack of focus on delivering a single version of truth. In my opinion, backed by a respectable number of disappointing projects, there's nothing like a single version of truth. In fact, enforcing it may do more evil than good.

Let's consider the most basic of all BI outcomes: sales.
More precisely let's talk about invoices. Nothing appears to be easier the defining the monthly invoiced amount; just sum all the invoices. Many senior managers will think this way and will be irritated by any further question, save being irritated as well by realizing that the accounting department provides different numbers than those in their sales reports.

How are credit notes supposed to be taken into account? Should they decrease the invoiced amount? All of them? And returns? What about those occasional invoices not issued upon a sale but for other services? Are all the products that produce a revenue relevant to sale performance (ex.. paid merchandising)? How many other particular cases lie inside a complex organization?

Accounting has its own standards to comply with, and knows exactly what to do with every document, so they're out of the game.
The other offices will likely have an answer for each of the questions above (and probably will be surprised that the answer is not obvious to everyone), but they'll all have different answers.

This is not misalignment, the point is that everyone has a different goal and measure her success on different metrics. So the sales manager should count in returns but not those caused by product quality issues; marketing will need to include the merchandising sales; production will be interested more in returns than anything else and the top management, by definition, should care about everything but should also be able to clearly discriminate each component. Everybody will have a different number, but it must be the number which actually measures her performance and must be confronted with.

Sales are only the most immediate example but the same happens in every other analysis topic. Costs, for example, are even more complex.

Our single version of truth paradigm so becomes something different.
What's crucial is the ability to relate each metric to the others and decompose the various effects which account for variations. There must always be a calculus through which the sales according to group A are transformed to the sales according to group B.
I often found useful sharing this schema with all the senior managers involved in the project. This extinguishes potential arguments from the beginning and promotes alignment among all the particular goals toward a common economic result.

This usually works well till the CEO abruptly asks for the sum of all invoices�

Enjoy

I think a lot

Admin — 2009-07-30T01:33:59+01:00

Those who know me personally will be surprised that, in the back of my brain, sometimes a thought passes by :o).

�

Now that the release 1.0 is approaching I'm frantically working on the new website, as the old one is really inadequate. Working on the site, I reviewed all the scripts from beginning and I realized one thing: communicating business intelligence is extremely hard.

�

Viney@rd (did you notice the creative part of the new name?) is a product aimed to partially fill a gap in the BI market, the segment of midrange Excel based BI applications. This gap is there because some players were acquired by enterprise level producers and others went out of business.
The segment is scarcely guarded not only for the "Excel" part but even for the "midrange".
Players consolidation had the side effect of creating few large, elephant sized, suites which are unfit for the midrange market. I'm not a capitalist shooter, this is perfectly natural from their perspective; being a large organization makes naturally think to large organizations. That's where the big money is, after all.
Some of these big players even question that a midrange market is possible. My experience tells that there is a lot of room there. Small or medium companies can run very complex businesses. Complex businesses are inherently sensitive to small variations. There must be a process in place to detect those variations.
If I should tell what the main feature of midrange market is, I'd say that the focus is not on the technical side (there are few technical challenges there) but on the process side. A good application for this market should feature the maximum freedom for the users to pursue and idea (I hope Viney@rd will do).

�

If you are still with me, it should be clear what the target is but how can I communicate with it?

�

There are two kind of people who infer on midrange market: end users and consultants.

�

The latter is the easier group to communicate with. Freelance consultants and small consulting firms usually know their business very well. They are fully capable of understanding my techno-speak and evaluate directly if Viney@rd is of interest.
I'm not neglecting them, I'll set up a partner program within the end of the year.

�

End users, on the contrary, are an entirely different matter. I think that the archetypical Viney@rd user will be the comptroller or the sales analyst; they may not be named in this way but these are the most Excel intensive functions in the company.

�

First: how can explain them the benefits of having a central data repository for their data if they are not used to BI? Every script I could come up with resembles those miracle knives ads, and I do not want to sound like a television sale. So I adopted a schema "if this is your problem, this is how you can solve it", but it produces rather long scripts and I'm afraid I'll lose them before the end.
If there is someone out there who can point me to a resource on the topic, I'll appreciate it.
Actually I produced an Unique Selling Proposition that sounds like this: "Connect your Scattered Spreadsheet Data into a Single Environment, and Do it Yourself!". I like it but does not outline clearly a benefit.

�

Second: how can I reach them? My company will be a totally web based company till I'll be able to afford a true commercial structure. I was almost unable to find a forum or a blog where controllers discuss about their job. It appears that extremely relevant business topics like management accounting, costing systems, customer profitability analysis, penetration coefficients etc. are not discussed on the Internet. This is rather unbelievable, so I suspect I'm making a big mistake somewhere.

�

I have the ambition to popularize business intelligence as far as it's practical, but I realized it's a tough job. OK, if it were easy, it would not be fun. Drop me a line on the subject, if you can.

�

Largest Vineyard Improvement Ever - new beta version

Admin — 2009-07-12T18:42:08+01:00

You did not hear from me for quite a long time. I did not give up, I'll never do! The point is that I had some personal issues which prevented me from working on Vineyard on a regular basis. My 9-6 job expanded as we feel the crisis as many other business do, so an extra effort is required to stay on the edge.

Anyway, I never stopped developing Vineyard, and now I make available a new version which features a lot of improvements. you can download it from here. It was clear I had to improve the user experience when I used Vineyard to replicate a small project I did with other tools. It was an enlightening experience. If the lacking of enterprise features is given for granted, loading data and replicating reports was not easy at all. This also confirmed a few beta testers comments, that, for sake of cleanliness, at first, I did not agree with.

So, at the end of the day, here are the new features.

Smart code finder

When loading data into models, often, Excel files containing descriptions, not codes, are provided. Editing those files by hand, even with the use of the assisted input feature, is really cumbersome. So I added a code finder feature which, from description "guesses" the code.
I made quite a lot of tests and, being a guess, it is not always perfect but it is rather accurate. Simply format the file in a Vineyard friendly format, leaving the code column blank. Paste it on a model members worksheet and select the "Add Codes" command.

Query Synchronization

It is almost guaranteed that you'll need to assemble in one table results coming from more than one query. So I added a feature to synchronize the two queries. That is, the result of the second query is printed on the worksheet in the same order as the first query, so each row contains figures that refer to the same dimensions combination. For those who are familiar with SQL, it's like a "join on the fly".
With this feature custom sorting and filtering are now possible. There are very few BI tools that feature query synchronization, and it is often a lacked feature.

Filters from cells

Now query filters values can be read from cells on the worksheet. This means that different queries with similar filters are guaranteed to have the same values. So, updating a workbook does not require updating every single query filter. Also, a query output can be used as the input for another query, thus introducing the basics for the sub query feature (another rare feature among BI tools).

Query Sequencing

Now that queries are no longer independent because of synchronization or filters from cells, sequencing their refresh is critical. Thus is now possible to sort the queries within the workbook, so that their massive refresh is made in the correct order.

Query importing

As you'll know, the query is saved within the workbook, but you'll likely need to embed a query in more than one workbook (a query with standard filters, for example). So now you can import a query from one or more workbooks, simply selecting the file(s).
Nothing prevents you from creating a workbook with all the standard queries required and share it among users.

I'm rather proud of this features set because makes Vineyard step out of first infancy, but I'm well aware that there is still a very long journey ahead.

Enjoy

Greatly improved query panel

Admin — 2009-06-11T22:57:47+01:00

Hello everybody.
Life has been hectic in the last days as I had few personal issues, but, most of all, I was sorting out a number of bugs from Vineyard and made the query actually usable.
I thank all those who had the patience to install Vineyard because the ModelDoc bug was really annoying.

The new query panel now features field sorting, hiding and sorting.
The filter panel has been improved and the long running queries now actually notify the user.

These features are now much more "industrial strength". Albeit Excel gives the user an unparalleled flexibility, a certain degree of good query formatting saves loads of work!

Feel free to download Vineyard from here. Simply install over the existing, the database is left untouched.

Enjoy

The project race

Admin — 2009-06-01T19:49:44+01:00

I've been a BI consultant since the beginning of my career. My experience ranges from small projects with a handful of users to large corporation level systems. I've been working in luxury offices with moquette and boiserie and in country settlements with the smell of the cattle coming in through the windows.

Despite this wide range of different environments, there is one element that accomunates them all: the ansiety to demonstrate that the project delivers actual value.

Better than ansiety, I'd say hurry, when not frenzy.
BI, opposed to transactional systems, is often hard to communicate when it is first introduced in a company environment.
From a commercial point of view it is relatively easy to rely on special effects and eye candy to sell the package. Costs often require the approval from people whose job is to run the company as a whole and does not have the specific knowledge required to anticipate what a BI project is really like. So it is almost unavoidable to create unjustified expectations. If I told you that the bulk of the job is ETL and data quality, would you have bought all the same? If you had known the effort required to define a common intelligence framework, a common language, common data rules, would you have signed the contract? THe answer is often a clear "no". You had kept going with your half baked reporting widgets.

So, what is the perceived way out? It is to think big and start small. Just start delivering a chunk of data, just start working with a group of users and bring them live as soon as you can. Just start delivering value and people will realize that this datawarehouse thing is helpful for sure.

But.

If development has been too fast, if users have not been properly involved, you are likely to commit a suicide that will permeate the entire project, if any remains. If results are unreliable and people did not buy in the project, you are doomed.
The initial stages of a BI project are a true minefield. An erp project fails at go-live, BI fails when you sit at the table for the second time. What's good is that, once the initial stage is succesfully over, everyone can breathe freely, and the following stages are easier to go through.

Of course, we BI professional do not need this sort of reassurance:We are able to see how the system will look like, to assess advantages and the ROI. It is rather easy if you know what you are talking about. The point is that few senior level managers have the culture to understand our point. They have no doubt about needing an erp to manage daily operations, but often do not get at all why they may need this datawarehouse thing. Usually a message like "consolidation will be easyer" or "you can have daily figures on your Blackberry" are effective but they subside as they are, giugno giving at best a partial image of the project. Nothing is more embarassing then a CEO asking the CIO for how the new consolidation system is going, and the CIO answers "what system?".

As usual there is no substitute for deep domain knowledge, understanding and trust. It is no longer cool to admit of knowing nothing about technology, a thorough knowledge of IT related issues should be part of a CEO knowledge no less than financials or marketing. Till that point our lives will keep being much more complex than necessary.

User experience is gettinge better and better ...

Admin — 2009-05-24T23:50:07+01:00

OK pals, here we are for our new beta release.
I'd label this a minor release too but these are all steps toward a better overall usability.

- I filled a wide hole in test data that made the same query return different results on pivot and Vineyard query.
- Sorting dimensions and measures in the query has been implemented by simple spin buttons. It was really a nightmare, before, to get the right order.
- List of values are now sortable, and the pleonastic click on the filter textbox has gone away (this is all for Brittany)
- Query renaming now does not lead to farcical results anymore.

I must remind again to those who subscribe to the beta preview program to leave their e-mail. The only way I have to identify you for sure is a mail exchange, I could not be able to provide you with the associated discount!

Now, please, download the new version from here.

enjoy

Minor release

Admin — 2009-05-19T01:53:54+01:00

I just released a new Beta version.

In the groove of a better user experience, it adds sorting to the list of values in the filters form. I also corrected a bunch of awful minor bugs.

To install, it, as usual, uninstalla the existing version and install the new. Nothing must be done to the database.

You can download it from here or from the usual download page.

Have fun!

New Beta Version � better user experience

Admin — 2009-05-10T23:11:51+01:00

This release comprises many of the suggestions I received from the early Beta testers.

The general point that emerged from the comments was quite clear. I devoted a lot of effort to the modeling tool and overlooked the Query Add-in. With this release I�m closing the gap. �The new version can be downloaded here.

�

I focused the effort on the modeling tool because I feel it is the key for Vineyard adoption.

As a good marketer, I thought about the sales channel element of the mix for a long time.

At first, I thought that the ideal target would have been executives dealing with data analysis and struggling with the spreadsheet hell. Actually they are, but Vineyard has an impassable drawback when marketing to them: it requires SQL server and such an installation is not within everyone�s reach. This means that the downloader should refer to its internal IT, thus delaying or stopping the entire process.

The second channel, which became clear to me only later, is the consulting channel. From a consulting perspective Vineyard has many strong points. First, it is rather inexpensive, leaving more customers� money for the project itself. It has a simple modeling tool to create star schemas to report against; even if you plan to build a classical ETL it is ideal to build and maintain a solution. Particular values and exception values are a breeze to load directly by Excel. Last, the overall �footprint� on users� systems is often negligible, letting you concentrate on delivering real value to the customer.

For both channels, the key element is the modeling tool for sure. There are a lot of data visualization tools, a lot of query tools, many Excel based, which are far better looking than Vineyard. None the less, none of these tools let you model your environment in the way Vineyard does.

I pushed this idea so far that I forgot that users must be able to retrieve their data in a comfortable fashion.

�

So here comes the first batch of updates for a better user experience.

Now a query can be duplicated and renamed directly from the query list. As a workbook has often many queries, differing only for few parameters (ex. YTD sales vs Last Year sales), this feature was mandatory. �Even renaming a query, with this feature, becomes absolutely necessary.

All the queries in a workbook can now be refreshed simply selecting them and clicking a command. There is no need to enter the edit form any more. This was also mandatory for the same reasons as above.

Last but not least, the inability to place a filter on an item not returned in the query was a huge limitation. The classical YTD required the year to be returned with the query. Place a dozen filters and you�ll obtain something unreadable. This feature has been implemented by introducing a �Hide� command to hide all the items which are not to be returned. Simply place a filter on the item than hide it.

�

Simply install this new version over the existing; the Beta databases are left unchanged, while the Alpha is not compatible.

�

Just a last advice to those who subscribed to the beta test program: please leave your e-mail! Some of you didn�t and now I do not have any mean to identify you and let you have the discount. The only way to identify you is having a valid e-mail address that I can check. As I explain in many places I�ll not spam you, I�ll only use it to verify your identity.

�

Now, please, download the new version from here.

�

Enjoy.

�

Our life on Earth

Admin — 2009-05-08T23:39:45+01:00

This is a Steve Jobs' speech. No matter what your think of him, listen to him, and have your kids listen to him. Enjoy