The latest news from the meaning blog

 

Rosetta Studio 3.3 reviewed

In Brief

What it does

Report automation platform which takes tabular output from almost any MR cross-tab program and transforms it into well-crafted PowerPoint presentations. Works with existing slide decks or will generate new ones selectively, directly from tables.

Supplier

ATP Canada

Our ratings

Score 4 out of 5Ease of use

Score 5 out of 5Compatibility with other software

Score 4 out of 5Value for money

Cost

Annual subscriptions from £5000 for a single user, £5,850 for 2 named users, £10,500 for 10. Concurrent pricing options also available. Prices include training, support and updates

Pros

  • Saves hours when converting tables into presentations
  • Greatly reduces the scope for errors when creating presentations
  • Shared templates reduce work and allow for a custom look for each client
  • Presentations can be updated with new data even after they have been modified in PowerPoint

Cons

  • No full preview without exporting to PowerPoint
  • No undo when editing or making changes
  • Windows only

In Depth

It’s a program that few profess to love, but PowerPoint shows little signs of yielding its iron fist over the boardroom presentation yet. Researchers often feel they are slaves to the god PowerPoint twice over, not just when presenting, but when preparing too due to the sheer monotonous drudgery of creating each slide by hand – slowly patting figures and graphs into shape, often against the pressure of the clock.

Rosetta Studio automates this process, and does it in a very efficient and comprehensive way. As a tool, it’s been around for over five years now. We reviewed version 1 back in 2005, when it was simple and straightforward to use, but fell short of doing all you needed it to. There wasn’t enough control over style and layout to create client-ready presentations, and this inevitably resulted in having to do a lot of finessing on the final output within PowerPoint to get the look right, and leaving you vulnerable to having to repeat all this work if you needed to re-run the data.

Improvements since then, culminating in version 3.3, have removed all these limitations. The range of capabilities has, quite simply, exploded. Pretty much any tabular input is now supported, either through built-in importers, or by using an upstream mapping utility. Within the tool, there is now fine control over every aspect of output, and a lot of attention has gone into providing ways to prevent anyone from every having to do anything more than once.

As an example, colouring, shading and chart options are all created and controlled within Rosetta and are not limited to what Excel or PowerPoint can produce. Colours can be linked with products or even applied automatically from the labels in your data for brand names, company names, countries and so on. It eliminates any fight you may have with PowerPoint showing different colours from the ones you had hoped for because of the clumsy way that PowerPoint templates and document themes interact. Instead. All of this is controlled, safely and predictably from within Rosetta, yet still it is standard PowerPoint that comes out of it.

A very powerful feature of the tool is the template. These takes a little time to master, but templates have the advantage that, once defined, they can be used across the whole organisation and shared easily among teams. Using templates, it takes just seconds to build charts from tables. Templates not only apply the styling, but work out what to pick out of the raw table – e.g. just the percentages or the frequencies, and not the total columns and rows.

Not everyone needs to become a template guru. It is not hard to modify them, or adapt them – but if you want to ensure a consistent look, and to control this, they can also be password protected against unwanted or unintentional changes.

There are now three modes of operation: generate, populate and update. In version 1 only generate was possible – this limited Rosetta to the “push” method of production, where you effectively created then exported a PowerPoint from scratch. Generate is ideal for ad hoc work, but not much help for any kind of continuous or large scale production.

Populate mode introduces the alternative “pull” method, where you can take an existing PowerPoint and link it to tables within Rosetta Studio by going through the PowerPoint document, stripping out the variable content and replacing it with a tag that will pull in the relevant data from Rosetta. Tags can pull rows, columns, individual cell, tables or sections of tables, and are to some extent smart – e.g. you can pull the first row, the second from last cell, or the column headed ‘females’. ATP’s implementation is delightfully simple, though it does take some effort to get your mind around the process. But it is ideal for large-scale report production on trackers, where many similar reports are produced on different cuts of the data, and the suite provides some batching and automation modules for power-users that go beyond the desktop interface.

Even more ingenious is the new “update mode” which achieves a kind of magic with PowerPoint. There is nothing to stop you from going in and making extensive changes to your presentation and still be able to update the PowerPoint, for example, because you had to remove a case and re-weight the data. Rosetta invisibly watermarks each chart it produces and uses these hidden tags to identify each value correctly and substitute the updated value. It’s very clever.

All this increased sophistication does come at a cost, however, as the price has nudged upwards, and so too is the investment you need to make in learning it.  This is not something you can pick up for the first time and guess your way through. ATP Canada encourages new all users to take their two-day training course before getting started, by including it ‘free’ within the licence fee. The program is reasonably intuitive, once you have got to grips with the fundamentals, though it’s a pity that you cannot get a better preview of what you are doing within the Rosetta interface – you only see it properly when you generate the PowerPoint.  If you find yourself going along the wrong track, it does not provide any real undo capability either, which is a pity.

Producing presentations is complex and without something like this, is very time-consuming. Speaking to actual users, it’s clear that users not only find learning it an investment work making, but that they soon wonder how they ever managed without it.

Customer viewpoint: Leger Marketing, Canada

Leger Marketing is a full service research company with nine offices across Canada and the United States. Here Christian Bourque, VP Research and Patrick Ryan, a Research Analyst at Leger Marketing, speak of their experiences with Rosetta Studio.

CB: “At the time we were looking for something, we felt most automation software was aimed at large trackers. About eighty per cent of our work is ad hoc. We needed something where the up-front time would be quite small. Rosetta Studio seemed to be better-designed for ad-hoc research, certainly at that time. “

PR: “Now, nearly every one of our quantitative projects runs through Rosetta at some stage.   We’d even use it for a four question omnibus study, where it is still faster than creating a report slide-by-slide. It means the bulk of our time is no longer taken up with the work of creating a report.  The focus is now on analysing and interpreting the data.”

CB: “Once you have spent a little bit of time devising your own templates, you will save 60 to 75 per cent of your time analysing the data. “

PR: “Something that would take four days or a week to put together is now taking us one or two days.   “

CB: “It not only saves the analyst time, but you also need to consider the quality review perspective. We used to do a line-by-line review. Now, because it is automated, this is no longer necessary. It’s a load off our shoulders. It means we can spend more time improving the quality of the insights.  We also find we can include more complex cuts of data in the report that we would not have had time to do, beforehand, like that little bit of extra multivariate analysis.”

PR: “Something we like a lot is the flexibility it gives you to try different things. You might be creating a set of graphs and you realise it could be better presented another way.  Now the hassle of changing your graphs or charts isn’t such a big deal. It takes you two seconds.

“It takes two days to learn, though the basics can be covered in a morning. It is fairly intuitive. We have a couple of reports where the analysis use the tagging. The interface is the same but the logic is different. You have get your mind around how to use and place tags, but once you have done one it is fine. It’s actually very simple.”

CB “We like the flexibility it provides from a look and feel point of view. We can have different templates for different companies. Many of our client have a corporate library of anything they generate, so when it circulates on the client side, it needs to look as if they it’s their document.

“This is something we introduced to add value, not to reduce staffing. It’s the nature of our business that you constantly have to be faster than the year before.  The demand on time is extreme. This is one of the ways we’ve been able to meet that challenge, while improving quality.  And the other major demand is for better insights, and this is one of the tools that allows us to do that.“

A version of this review first appeared in Research, the magazine of the Market Research Society, February 2010, Issue 524

mTABview previewed

In Brief

What it does

Report automation software to convert mTAB cross-tab analysis directly to PowerPoint presentations containing tables and charts, with push and pull technology to allow for easy update for trackers or new releases of data.

Supplier

PAI & Gamma Associates

Our ratings

Score 4.5 out of 5Ease of use

Score 4.5 out of 5Compatibility with other software

Score 5 out of 5Value for money

Cost

Available as an add-on to mTABweb at £500 per required user.

Pros

  • All-in-one process for analysing data and creating PowerPoint slide
  • When adding new waves, overlay tool detects any differences and helps you to reconcile them

Cons

  • Charts limited to histograms – no pie chart
  • Cannot create your own templates yet
  • Windows only

In Depth

In the Autumn, Gamma will be releasing mTABview, an add-on to mTABweb that will allow users to build PowerPoint slides directly from their mTAB databases. We were given an unreleased ‘beta-test’ version to review.

mTABview is also a web-based program, but is Windows only. Rather than adding more features directly into the mTABweb interface, mTABview works in its own browser window. Logging in, you start by selecting the template you wish to use, as you would with PowerPoint. It works in tandem with mTABweb, pulling its data directly from the same database structures, and driving the mTAB analysis engine. This gives it an advantage over other PowerPoint creators like E-Tabs and Rosetta Studio, which can only work with static outputs, as you can refine and reanalyse you data as you go. Although the tool produces PowerPoint, you do not work in PowerPoint – it produces the PPT file when you are finished. However, as you build your deck within mTABview, you work with completely life-like previews of the actual PowerPoint slides. It offers a range of different slide types to work with: a title slide, a section header slide and information slides organised into one or two columns.

You might start by setting up the title slide. The template will provide defaults for font, size, or colour, to save time and achieve consistency. Any of these options can be overridden easily and intuitively, but regrettably in version 1, there is no way to save your modifications as a new template or to create your own – you would need to get Gamma to set one up for you.

Next, you define the slides you wish to see, and choose a one- or two-column layout into which you add your table or chart. As soon as you pick either, mTABview hands you over to mTABweb to build your table. You are then back in the familiar Filofax view to assemble your table and run the analysis. Hit the button to run the table, and it will post the results into the mTABview window.

Anyone preparing presentations will know that much of the time is spent in picking the right subset of data to show – combining categories or omitting them, suppressing either frequencies or percentages, switching off total columns and rows and hiding ‘Others’ and ‘Don’t know’ categories. All of this is made pretty effortless by mTABview. You pick the figures to display by clicking on the relevant column and row headers and the selected portions are clearly highlighted. It also lets you choose whether you work with frequencies or percentages. If you need to combine categories, then you can post the analysis back to mTABweb. It cleverly stores the original query you used to generate the analysis, so at any time – even months later – you can go back and recreate the table from the source data. From within mTAB, you can easily perform any recodes or combinations you need, and it will always post the results straight back into the slide you are working on.

Charts are just as simple. It does not use the PowerPoint chart engine to build the charts, but ingeniously, creates native PowerPoint charts in a two-stage process. When you export your finished deck to PowerPoint, you can use a small mTABview plug-in to pull in proper, editable PowerPoint objects, from the charts and tables created in mTABview.

Sadly, the current the range of charts is very limited – there are histograms in various forms and that’s it. No pie charts, no correspondance maps (even though mTAB will produce them), and no automated means to highlight significant data. These omissions are likely to disappoint early adopters when the program is released, though Gamma should be able to address them subsequently, now the building blocks are in place.

There is also excellent support for trackers, as there is in mTAB. Here, it uses the concept of an overlay. You take a previously saved presentation, and overlay a new data source on it. The program rattles through and reports all of the items where there are changes, not just in the labels, but in the underlying data structures. You can then click through these and decide how to resolve them there and then. Updating PowerPoint decks manually can take as long as creating them in the first place – but use this, and the prospect of a tedious day or two of work and careful checking vanishes into an hour or less.  If Gamma can grow mTABview to offer more flexibility in its outputs, this software is potential killer app material.

A version of this review first appeared in Research, the magazine of the Market Research Society, August 2008, Issue 506