5 Reasons Buyers Are Pleading “No More Spreadsheets!”
Yes, a spreadsheet is an indispensable tool for creating small catalogs of electronics parts for buyers to review. But buyers have become sensitive to its limitations, which are painfully evident in various non-trivial situations. Ideally, a search for electronic parts yields a quick, comprehensive, and accurate result. Suppliers are not able to meet those criteria when buyers are forced to manually review a set of spreadsheets, aggregate and normalize items of interest, and compare the final results. Discerning customers are abandoning this outdated approach to parts searching at every opportunity. Recent advances in automation and fluctuations in market conditions have only served to encourage the shift to a more efficient approach. The reasons for such a transition are clear:
1. Inaccurate data.
There are many examples of significant, costly mistakes being made in spreadsheets across industries. As many as 90% of spreadsheets contain errors. Less publicized are the small unnoticed errors, invisible for all practical purposes and under minimal scrutiny. A manual search of more than one spreadsheet virtually ensures you’ll encounter errors. The results of such errors affect parts delivery, support efforts, and overall operational cost.
2. Spreadsheet resource limits.
The electronics industry produces an astronomical number of electronic parts and related variants. A single spreadsheet can only handle about one million rows of data or about 65,000 embedded hyperlinks. Even when a catalog can conform to these limits, spreadsheets easily reach a size that makes them unwieldy to search and review. A supplier using an over-powered machine can create a document that is not even fully accessible on a buyer’s machine due to memory limits.
3. Stale data.
The shelf life of a downloaded spreadsheet isn’t always apparent to an end user. By the time the data is accessed, the information of interest—including availability and price—can be grossly outdated and inaccurate. In addition, a spreadsheet can be copied and moved to different locations which further obscures the exact time it was originally downloaded. Spreadsheets are inherently decentralized stores of data, making them extremely flexible and easy to access. Unfortunately, users must then sacrifice the typical control expected when entering into substantial financial transactions.
4. Difficultly in making comparisons.
The malleable nature of spreadsheets makes them internally inconsistent. They are manually constructed by individuals with few standards or mechanisms to ensure consistency among different documents. These challenges make comparisons within and between spreadsheets incredibly difficult.
The minor inconsistencies that often exist within spreadsheets makes understanding the data itself arduous. Formatting differences or text-based descriptions of quantitative information with varying units forces a “normalization” exercise on the part of the end user to make proper comparisons between rows.
In addition to the internal challenges, each spreadsheet has a different format. Additional mental overhead is required to scroll back and forth, filter, convert units, and ensure “apples-to-apples” comparisons between spreadsheets. This task is both extremely time consuming and error prone.
The “Find” function has a number of subtle options. Scoped searches control whether a given search occurs within a single sheet or all sheets contained in a workbook. Searches can be case sensitive or ignore the case of the text in question. A search can require that the entire cell matches or can be limited to finding a partial match in a given cell. Filtered values further complicate the situation. An unwary user can do a search that suggests data is not available when it actually is due to the search not being performed with the intended options.
5. Accidental updates.
While clicking and scrolling within a spreadsheet to obtain information, a single mistaken keystroke can result in an inadvertent update to the data. Even if the spreadsheet was perfectly accurate when initially acquired, it can have changes applied that are not evident to the end user. Again, the very flexibility that makes a spreadsheet useful makes it unsuitable for controlled, auditable transactions.
These limits are well understood and have resulted in the rise of enterprise resource planning software (ERPs) and industry specific systems that address the problems described. Tony Powell, CTO at Orbweaver, makes a compelling case to avoid searching spreadsheets and use dedicated catalog search systems:
“A computer can search hundreds or thousands of suppliers and millions of parts in a second or less, and a human can search a handful of suppliers in an hour. Studies show that our platform reduces the human per-line procurement cost from around $25 per line item to around .25 cents.”
Dedicated systems address the limits described above utilizing fundamentally different technology. Data validation routines and well-designed user interfaces prevent the ingestion of inaccurate data. Modern SQL and NoSQL databases are not hindered by the size and memory limitations of spreadsheets. Data is updated in a central location in a timely manner. Searches are consistently implemented and provide instant, uniform results. Best of all, access control is built into such systems, preventing accidental, untrackable modifications from occurring. They are fundamentally the right tool for the job used by discerning buyers.
