3 ms·
Can’t say I’m too concerned. Every project I’ve ever worked on that outsourced something ended in failure or a major decline in quality. We used to need 3 accou
by diogenescynic 4y ago
Can’t say I’m too concerned. Every project I’ve ever worked on that outsourced something ended in failure or a major decline in quality. We used to need 3 accountants in the US, now we have an army of BPO accountants and none of them know what they’re doing and things take longer than ever. Outsourcing doesn’t work. It’s not even clear it costs less when you factor in the decline in quality, delays, and extra work needed by everyone else to monitor the outsourced work. The juice isn’t worth the squeeze.
- tablespoon 4y ago> Can’t say I’m too concerned. Every project I’ve ever worked on that outsourced something ended in failure or a major decline in quality. We used to need 3 accountants in the US, now we have an army of BPO accountants and none of them know what they’re doing and things take longer than ever. Honestly, I don't think corporate decision-makers really care about those metrics. > It’s not even clear it costs less when you factor in the decline in quality, delays, and extra work needed by everyone else to monitor the outsourced work. The juice isn’t worth the squeeze. So it's six of one, half a dozen of the other then? That's not a very persuasive argument against outsourcing.
- taylodl 4y agoIt’s not even clear it costs less when you factor in the decline in quality, delays, and extra work needed by everyone else to monitor the outsourced work. The juice isn’t worth the squeeze. How do you read that as six one way, half a dozen the other? Outsourcing often leads to lower quality, delays (which has business opportunity costs) and extra expense to monitor and review the work. Instead of being faster, cheaper, better outsourcing is slower, more expensive, and worse. Why would anybody choose that?
- tablespoon 4y ago>> It’s not even clear it costs less when you factor in the decline in quality, delays, and extra work needed by everyone else to monitor the outsourced work. The juice isn’t worth the squeeze. >> So it's six of one, half a dozen of the other then? That's not a very persuasive argument against outsourcing. > How do you read that as six one way, half a dozen the other? If the costs aren't significantly higher than in-sourced work in ways quantifiable in dollars, then I don't think it's going to matter much to business leaders. In other words, they just don't plain care about the things we care about. If the thing gets done crappily and painfully, it's still done, and that's fine with them. Also, I don't think businesses are logical or rigorous places. IMHO, it's par for the course to "save" money by shifting costs into more difficult to quantify areas. For instance: cut desktop support, and the costs move into the groups being supported, so long as the complaints aren't loud enough to reach up through many layers of management, you have yourself a "win".