In the old days -- five years ago, to be exact -- a financial spreadsheet was manufactured out of actual paper and ink. It often measured two feet across and was unlikely to remain long on top of a desk without having coffee spilled on it. Against its typically dull-green Eye-Ease background were crosshatched some 7,500 spaces into which accountants, bookkeepers, treasurers, and business planners squeezed their numbers, tiny oblong by tiny oblong, pencil by resharpened pencil. It was white-collar sweatshop work; a manager performing a five-year projection involving even a handful of products could count on many weekends in the office and wastebaskets full of jettisoned calculations.
Happily, that kind of tedium drew to an end with the dawning of the age of the electronic spreadsheet in October 1979, the month that VisiCalc, the first of the new breed, was shipped by Software Arts Inc. Today there are some 150 personal, or desktop, makes of microcomputers capable of running financial-planning programs. Since there are an estimated five dozen such electronic spreadsheets, the permutations of hardware and software themselves are enough to fill any one of them to overflowing.
But now the headlong rush of technology is antiquating even those 60 programs and the concomitant graphing and database software that followed on their heels, plugging into microcomputers disk by disk. Last July, Context Management Systems, of Torrance, Calif., released Context MBA, a complex, highly sophisticated program that integrates five separate applications on a single disk. Without changing disks, a user now can switch among business modeling, graph-making, word processing, telecommunications, and database functions with only a keystroke or two and can rapidly plug information utilized in one mode into any other mode. Context claims that MBA, which retails at $695 for the IBM version and $795 for the Hewlett-Packard version, is "among the largest microcomputer programs ever written." The idea behind such almost-unimaginable -- to personal computing, anyway -- immensity is that the speed of calculation and the handiness of formatting information in different ways will fire up management creativity and make desktops more productive.
In a classically free-enterprise instance of inventors separately responding to essentially the same inspiration -- the mother of invention in this case being a financial-planning market estimated to reach some $300 million by 1985 -- the publication of MBA was followed within a few months by a similarly integrated management tool created cross-country by Lotus Development Corp. in Cambridge, Mass. Called 1-2-3 (both because it is as fast as that, explains Lotus, and because it combines three functions: modeling, graphics and database), the $495 program, Lotus claims, "will advance computer productivity in the same order of magnitude as [did] the introduction of the electronic spreadsheet."
If that prediction means that five dozen more integrated, multitask, executive-productivity programs are shortly to explode on the desktop scene, ranagers are in for some special decision-making headaches. To appreciate the full dimensions and various nuances of MBA and 1-2-3, and to be able to distinguish the important differences between them requires weeks of concentrated study even by a computer-literate manager. But the fuse may be lit already. Apple's new Lisa runs software that is not unlike MBA and 1-2-3 (it is not quite so readily integrated but in many ways is more sophisticated), Microsoft Inc has made its spreadsheet program considerably more far-reaching; and VisiCorp has announced (but as of this writing has not released) a new, "mouse"-driven generation of business software that, according to the company, "lets any personal computer user easily work with a number of application products at one time." These are just a few instances of what already promises to be another inundation of financial-management computerized tools, a gathering that is sure to be as diverse -- and therefore as confusing -- as the first wave.
To help sort out the factors and demonstrate how such programs might be applied to business operations, INC. has studied 1-2-3 and Context MBA, the only two fully integrated worksheet programs available in the desktop market this spring. 1-2-3, which requires a minimum of 192 kilobytes of random access memory (RAM), was run on an IBM Personal Computer, Context MBA, which eats up 256K, was run on a Hewlett-Packard 9816. This report is not meant to be an opinionated review so much as a simplified presentation of the applications of the programs, together with a few elements of comparison.
Essentially, a worksheet program saves time and erasers -- lots of both. Just as on a paper ledger sheet, virtually any aspect of a business can be modeled, by plugging raw data into a computer, a manager can review current conditions and can try out a variety of "what if" situations, eventually determining, at least so far as such statistical exercises point the way, the most productive strategies.
For example, by setting up a table of accounts receivable, accounts payable, interest rates, and so on, cash flow can be analyzed. By comparing results over the last five years, say, profit margin and profitability can be examined and weaknesses determined. Operating models of balance sheets for a company's divisions can be created and consolidated. Bottom line results can be studied in terms of differing inflation rates or accounting procedures. And so on through the dynamics of modern enterprise. Indeed, so extensive are the applications of an integrated spreadsheet, with its ability to accumulate data freely from many sources, that the capacity of both MBA and 1-2-3 are themselves almost endless.
It so happens that 1-2-3's capacity is more endless than MBA's. Its spreadsheet is made up of 2,048 rows by 256 columns -- the largest electronic spreadsheet available, claims Lotus MBA's is 999 rows by 95 columns. But much of 1-2-3's vastness is unchartable land: Because of memory limitations, not all of it is usable at once. The entire spreadsheet of MBA, however, can be utilized at once. Each of Lotus's 524,288 and Context's 94,905 areas thus defined in its basic spreadsheet format is called a cell, and into each cell can be placed a value (or other material) upon which the computer will, if called upon, perform a calculation. In many instances, that calculation can be accomplished literally in the blink of an eye. To appreciate the significance of this modern miracle, consider that the same amount of 1-2-3 work space on a typical 25-column by 41-row ledger leaf would necessitate a pad of 511 pages.
The MBA spreadsheet contains only 18% of the amount of cells as the 1-2-3, but it makes up in cell capacity the ground thus given away in sheer numbers. In the modeling, or numeric-analysis mode, each MBA cell can hold 502 characters versus 1-2-3's 240. Unless you are placing a satellite in precise orbit, a given value won't require that much room, but a formula or graph instruction might (see illustrations). In its word processing and database modes, an individual MBA cell can hold about 8,000 characters, or about three 55-line pages. And if that isn't enough for a stern interoffice memo, it can then spill over into a contiguous cell. 1-2-3 lacks a true word-processing function altogether. (Lotus promises that its next revision -- or "rev" -- will contain a word-processing program.)
A group of cells comprise a range, within which individual entries relate to each other. To establish this tabular relationship, a cell can be equipped with a formula for the operations to be performed on it. The user can either write a formula, or call up one of the many financial, statistical, mathematical, or logical functions built into the program. Most are complex, such as internal rate of return and analysis of variants. A manager could take cost-of-goods-sold as a percentage of gross sales by programming an empty cell to add all cells containing fixed and running costs and divide that sum by the cell containing total gross sales. But he would not have to rewrite so laborious a formula each time he needed it. In addition to its ability to move effortlessly from database information to financial modeling to graphic displays (and, in MBA's case, to call on mainframe computers or telephone databases for data), one of the most powerful aspects of an electronic worksheet is the ability to copy, or replicate, formulas relative to their position within a range. The user just tells the machine where to put the same formula; the program figures out what other cells are relatively or absolutely involved in the new position.
That is when a manager gets to show his stuff. Once a given statistical situation is formatted, values can be changed at will to observe the effect a different argument, as it is called, has on the whole (a drop in six-month T-bill interest on a bond portfolio, for example, or a change in inflation rates on real-dollar revenues). 1-2-3 instantly recalculates all the formulas in natural order (that is, it will wait to use a formula that is dependent on the changed values in other formulas), but the interrelationships can be altered by the user if necessary. Both MBA and 1-2-3 provide for manual recalculation, but, less flexibly, in MBA the sequence can be changed only row by row or column by column. On the other hand, MBA conveniently allows a given range to be defined and recalculated without disturbing the rest of the entries outside the range -- a process that conserves operating time on slower-running MBA.
Such discrepancies -- many of them simply annoying or minor -- occur in abundance. Other disparities loom larger. In particular (as of this writing) is MBA's quite workable, although rudimentary, word-processing function, in which documents such as short reports can be created. In them, data from the database, modeling, or graphing functions can be incorporated. Another major difference is that 1-2-3's graphs cannot share the same screen with tables; either the text must be blanked and the graph displayed in its place, or a separate monitor must be connected to the system. More important in this area, MBA can split its screen into four interactive parts, while 1-2-3 divides only in two, either horizontally or vertically. An advantage to the MBA split-screen setup is that it allows immediate "what-if" viewing in key sectors -- three graphs, for example, changing as new values are plugged into a table displayed in the fourth window. Furthermore, MBA can print a graph directly from the screen, with 1-2-3, a graph must be stored on the disk first and the program disk changed.
1-2-3 offers a powerful alternative allowing a user to alter program operations himself. Called macro programming, the ability to enter sequences of keystrokes in a succession of cells and execute complicated operations via a single command lets a user create his or her own menu such that anyone could run it without training.
But as each publisher discovers what the other has ingeniously stuffed into its package and as it receives feedback from users, its program gets upgraded. One critical difference has been the comparative response time of each. No manager wants to sit around paring his nails while some microcomputer chip is scratching its head inside a box. In speed, 1-2-3 has excelled throughout on the IBM PC -- the only operating system for which it has been written so far. Response time in MBA's upgraded version for the IBM PC has been enhanced to the point where it is quite tolerable, although it continues to lag behind its rival. On the faster 16/32-bit Hewlett-Packard 9816, however, MBA performs even extensive operations comparably fast.
Earlier this year, Context added a telecommunications mode to its package, leaving Lotus eating dust in the data-acquisition arena, at least for the time being. The program is extremely flexible. Information can be transmitted to or be gathered from a host computer and placed automatically into an MBA file or work space. Data also can be received (at 300 or 1,200 baud) from such telephone-accessed databases as Dow Jones News/Retrieval Service via an auto-dial capability. With a split screen, one window can be kept in the communications mode ready to take a call while the user works on material in the other windows. Then the data can be received in its entirely within a designated cell.
In file management -- the storing and retrieving of data from disk -- the two programs differ to a significant degree. MBA terms its data disk a "volume"; the volume is broken down into files, within which are "documents" -- worksheets relating to that file name. 1-2-3 stores information only in files one worksheet at a time, without differentiation. Both display file names when called; MBA also reviews document names within a file simply by scrolling. 1-2-3 displays how much storage space is still available on the disk (at the peril of not being able to fit in a spreadsheet, MBA does not), how much has been used for each file, and the time and date it was created -- none of which MBA discloses. When a file is saved in either program, all commands, formulas, and formatting remain. A retrieved file will come up just as when exited, cursor in its original position. 1-2-3 can import data from other files that are compatible with the PC operating system, and, among other subtle provisions, it can bring all or part of a saved file into a current file at a specific location.
It can be seen by now that the two programs -- each ingenious in certain ways and clumsy in others, deep at one spot but shallow at another -- are at variance at just about every turn, many of which have barely been mentioned here. But all told, both are currently powerful tools for management decision-making and magical massagers of statistics and strategies with capabilities that were virtually unthinkable five years ago. At the moment, it seems fair to say that 1-2-3 is the more clever and more easily used for what it does but that MBA does more. On the one hand, 1-2-3 may catch up with certain glaring omissions, in its next or subsequent revisions. On the other hand, MBA may equally refine its performance and add other ingredients. As long as the competition remains keen and new versions of these two programs keep coming -- and they are bound to, since other software houses already are preparing for the fray -- if a manager goes wrong with either, he or she had better look to his own MBA -- as quickly as 1,2,3.
Creating a rudimentary, fictional, and admittedly improbable book-publishing business, INC. has filled in some of the cells of an MBA spreadsheet on a Hewlett-Packard Model 9816 desktop microcomputer whose video display appears here. Like seating in a theater, each cell is identified according to row and column. The first cell, in the left-hand corner of the sheet, is A1 -- a location known as its "address." The area of each cell is adjustable -- by width in 1-2-3, and by both width and height in MBA. In this example, INC. has chosen to narrow the width of column A, merely on the aesthetic grounds of thus being able to automatically put some space between the titles in Column B and the numbers of rows that run down the left-hand side of the sheet. A user to whom such details are important will have an awkward time with page design. To gain the same effect as programmed, INC. could have elected to use Column A for book titles, placing entries in the center or flush right by punching a succession of five keys after each insertion, 1-2-3 also requires five strokes for cell-entry alignment. But in other instances, repetition is well accommodated; for example, also for its eye-pleasing effect, INC. has placed a dashed line between the column labels and underlying information. This can be done in either program by typing hyphens in one cell and telling the computer to repeat this entry in every other cell to a certain destination.
Many options offered by one program are not offered by the other. For example, to be able to spell out each book title in its entirety, Column B has been widened from its usual (or "default") setting of 7 characters (1-2-3's default column width is 9) to 26 characters. Had INC wanted to conceal this data, it could have compressed the MBA column down to 0 (the data thus remains inside the computer but is held back from view); a 1-2-3 column could not be compressed. But entries in a 1-2-3 cell can be protected from inadvertent change -- locked in, as it were -- whereas MBA has no such provision. So that a user can quickly identify the entries in Column D as fixed prices, INC. has placed a dollar sign in front of each amount -- again, a repetitive process requiring several keystrokes. Far more practically, 1-2-3 offers an easy-to-use, built-in currency format (a dollar sign, commas, and two decimals) thnt need be set only once for individual columns. 1-2-3 has built-in clock and date functions, MBA does not. And so on down a long -- but for the most part unimportant -- list.
At the top of the screen is a separate two-line information area (1-2-3 has three lines) that tells what file is being used in what mode, which cell is involved, what the contents of the cell are, what command has been issued by the user, what the computer's response to it is, and how much room is left in that particular file (or, in MBA's case, "folder"). Entries are typed out or edited here before or after being placed into a cell. In this screen the operutor has asked the computer to delete two entire rows (the "/D2R" at the top left). Since so doing could obliterate substantial amounts of data, the program is making sure that it is really the user's intention by querying, "Ok to do large operation? [Yes]/[No]" -- one of several such precautions that MBA frequently takes.
Both MBA and 1-2-3 can be formatted to round off values to fixed decimal places. And both store the exact value in memory for precise future reference if needed. Note that because most of the entries in Column H do not fit on the screen in their entirety, the MBA program automatically blanks them to prevent confusion, 1-2-3 simply omits a column from the display if it won't fit in its entirety. The contents of this screen can be printed by both MBA and 1-2-3.
INC. has split the screen into two windows -- a mode that both programs allow. (More powerfully, however, MBA can be split into three or four as well.) This allows the book titles at left to remain in position while other columns are brought alongside. In both programs, left-hand and top labels can be cemented in position if desired, so that they can't be scrolled off the screen. But in summing Column J, which happens to be projected gross sales for 1985 (H is '83, I is '84), INC. has chosen to omit the column labels in order to be able to fit all 17 books on the screen at once for reference. The cursor is at cell J23, containing the addition formula -- ? SUM(J5 . . . J21 -- shown at upper left. INC. has formatted these projected sales to display no decimal points and to be rounded off to the nearest dollar. Formulas have also been implanted in H5, 15, and J5 to multiply projected number of books sold (E5, F5, and G5, not shown) by retail price (D5, not shown). These formulas were then reproduced, or replicated, automatically to do the same calculations relative to each title.
A three-year summary of INC.'s overly optimistic forecast for brisk book sales has been constructed, based on extracts of work on the spreadsheet. Fixed and variable costs are plugged in and the result brought down to profit before taxes. Note the formula shown in upper left regarding the cell where the cursor (the white rectangle at bottom center) is located. In the second window alongside, net and gross profit have been plotted contiguously in a simple bar graph. The MBA program is able to shoehorn a graph in any window; to fit in the appointed space here, it appears narrow.
"What if" questions -- suppose binding costs double, or inflation affects 1985's manufacturing costs -- may then be asked of the table at the left, the results are instantly recalculated and redrawn on the graph. More complex situations may be similarly portrayed. For example, a decision might be made by "what if" analysis on whether it is better to print more books at a lower unit cost gained by quantity discount, pay for storing them, and sell them off over a longer term. Cost of goods sold would be cut and the year's net enhanced, but a manager would have to be wary of balancing this expedient against future inventory write-downs.
Both programs have the capacity to draw high-resolution graphs in black-and-white for MBA and in either black-and-white or color for 1-2-3 from multiple sets of data. With five basic types -- bar, stacked bar, line, pie, and XY (in which both axes are assigned a numerical scale -- a salesperson's salary, say, plotted against his gross production) -- 1-2-3 falls short of MBA's nine -- pie, bar, stacked bar, perspective bar, multiple bar, line, area, scatter, and high-low-close. MBA can overlay one type of graph on another. But 1-2-3's type fonts and grid selection are more extensive. Unlike 1-2-3's, MBA's graphs, such as the two shown here, can be printed directly from the screen, 1-2-3 requires storage to disk. Like an electronic interior decorator, the program itself can select the patterns, change type styles (in 1-2-3 only), or add grids (shown in the three-bar chart). It also automatically selects the Y-axis scale, lays out the bars according to how many have to be fit in, and, in the case of the pie chart, puts in the legends and performs the percentage breakdowns. The line directly above the simple pie chart illustrates a graph-making procedure. First, the computer is told what type of chart to construct -- "? PLOT(PIE)" -- then where the data is that it is to use: here, the row compriring cells Q23 through S23. Next, the chart can be titled at the top, bottom, and/or side, and the X-axis labeled. The instruction for the three-bar chart shows the directions for finding the X-axis labels and for a horizontal grid. 1-2-3 allows two-line titles, MBA does not. Note the computer query "Expand cell to full window?" By answering yes, a cell cited only as "graph" on the spreadsheet (see cells C42 and C43, bottom, page 108) blossoms forth into full linear array.
When formatted for database operations (horizontal rows are records, vertical columns are fields), both 1-2-3 and MBA have extraordinary depth. 1-2-3 can handle up to 2,047 records and up to 256 fields with the total number being limited by the computer's memory MBA has even greater record capacity -- 16,000 -- due to its ability to retrieve data from the disk but is limited by spreadsheet width to 95 fields; each field, however, can contain 8,000 characters. Both can perform Boolean logic operations (and or, greater than/less than, and so on) Here, one of MBA's stronger features: its 8,000-character cell capacity. The cursor in the resume field is pointed at an underproductive employee: one Robert Mamis. The database may be queried through 20 fields using up to six criteria. If Mr. Mamis's case is being studied, the operator can call his resume, filed in cell G4, to the screen through an operation called cell expansion. Other MBA attributes include the ability to customize formats, such that the spreadsheet configuration need not be adhered to, and an alphabetized indexing feature that speeds up disk retrieval. 1-2-3 enables an automated data table that responds to "what if" calculations on two lists and calculates the result on a single formula -- profit margin per region of country in regard to G&A allocations, for a simple example.
Both MBA and 1-2-3 can split a screen and perform operations in one sector that affect another. The windows can be either synchronized (such that scrolling one also scrolls the other, keeping columns and rows in constant relationship) or unsynchronized. Here, in MBA, two different graphs are responding to a typical problem in cash flows from investment options. Column B contains the initial cash outlay (and hence is negative; 1-2-3 permits formatting negative quantities within parentheses, as well as with minus signs). Those to the right are the resulting positive or negative cash flows for five periods. Both programs contain built-in formulas for, among many others, net present value and internal rate of return; only MBA has modified internal rate of return (a formula that assumes that borrowing is to be done at the "safe" rate and reinvestment at the "risk" rate; in this illustration the discount rate, or cost of the invested money, is 18%). The ability to split a screen is also helpful when writing formulas from one display that apply to others.
1-2-3 can't work both text and graphs on a single screen. Either the text must be blanked out and the graph called up, or a separate monitor must be used. In MBA's Version 2.0, the screen can be printed as shown, in another MBA revision word-processing text can be edited within a window.
At any phase of operation, 1-2-3's "Help" can be summoned, temporarily substituting a review of operation options for the screen page being worked on. One of some 200 pages of on-screen instructions is shown here, photographed on an IBM Personal Computer CRT. Each menu option has its own fully illustrated documentation that enables a novice to proceed without recourse to the printed documentation. By moving the cursor to the needed subject matter, a user can track down pertinent elements of operation. In contrast, MBA's on-screen support is minimal, designed mainly to reinforce tasks already learned from its tutorial. In spreadsheet operation, 1-2-3 menu options are displayed and briefly explained within the top information area and are selected either by cursor or key.
Unlike MBA, 1-2-3 makes copious use of its reverse video display. Here, a range that is to undergo recalculation has been "pointed at" by moving the cursor from cell to cell, thus causing the area involved to become unmistakably highlighted on the screen. As with MBA, range addresses may also be typed directly, although Lotus claims users prefer to "point." The 1-2-3 spreadsheet axes remain highlighted in reverse video throughout. 1-2-3's cursor encompasses the entire width of a cell, reverse-videoing its contents. Two lines' worth of the entire content or formula residing in the cell is always displayed in the upper-left information area along with the cell address, ready for editing. 1-2-3's "label" mode facilitates typing in titles without fussy formatting; lacking such a mode, MBA requires laboriously filling each individual cell with contiguous material.