4 ms·

Disclaimer: I work for an Accenture subsidiary. First, just in point of fact, I have personally been part of lots of successful deliveries. Both Facebook and

Disclaimer: I work for an Accenture subsidiary. First, just in point of fact, I have personally been part of lots of successful deliveries. Both Facebook and Google spent $1 billion with Accenture last year. Like all large IT consultancies, they're defined by normal distribution - they have high and low performers, pockets of competence and incompetence, big wins and big misses.

osrec 8y ago
I hired Accenture while I was an MD at an investment bank for a data migration project. I was convinced by another MD who used to work for Accenture. Anyway, we spent about £15m over a year, with literally no results to show for it. I was less than impressed - in fact, I was furious at the gross misrepresentation of the status of the project by the partner. My own experience leads me to believe that the distribution of talent is most certainly non-normal - perhaps skew-normal, with most of the weight in the less-competent zone.
saiya-jin 8y ago
For what Accenture (and probably other similar companies) delivers, for given price point, i don't get how they are still in business. Results are always super costly, and often meh or okayish at best. The variability of quality of people is simply too big and this is completely on Accenture's side. Yes some locations provide reliably good quality, but Accenture doesn't have the same standards for hiring across the globe. Seen first hand, couple of years ago batch of folks from central/east europe was hired for long term cooperation, out of 20 maybe 2-3 were worth their salt. For rest, firing them on spot would be more time-saving than baby-sitting extremely junior clueless unmotivated crowd.
1PlayerOne 8y ago
Can you tell us, in general terms, some projects that were successful engagement with Accenture?
- kthejoker2 8y ago
  Sure. I work in data and AI, so I'll focus on those as I know them best. We stood up and currently manage the trading platform, integration layer, data lake and analytics infrastructure, event messaging, on-demand machine learning compute provisioning and model management, API middleware, front end web app, and about 500 various service touchpoints therein of the largest oil and gas upstream finance trading desk in the world. We've developed a ton of standard demand planning and SKU-level pricing forecasting pipelines, ingest, storage, processing, model training and deployment, and consumption. In some ways that's our bread and butter as you can re-use 75-85% of the IP and focus a lot of energy on feature engineering and edge cases. We do a ton of IoT work in the mining and energy industries. Mostly (again) developing the infrastructure to ingest and process that data, building ML models to predict anomalies, production trends, workflow to connect that to SAP or some other ERP system for work orders, remote operating center dashboards, etc. We do a lot of HSE initiatives. One cool one we did recently was use some Zebra beacons and some computer vision to identify people entering dynamic "red flag" geofences (think mobile equipment, forklifts, cranes, wrecking balls, etc.) on field sites so we can instantly stop work. Anyway, I'm the first to admit Accenture's a mixed bag, but one thing I've learned since coming to work (under/for) them is large-scale delivery is very, very, very hard to 1) scope, 2) de-risk, and 3) price.

osrec 8y ago

I hired Accenture while I was an MD at an investment bank for a data migration project. I was convinced by another MD who used to work for Accenture. Anyway, we spent about £15m over a year, with literally no results to show for it. I was less than impressed - in fact, I was furious at the gross misrepresentation of the status of the project by the partner. My own experience leads me to believe that the distribution of talent is most certainly non-normal - perhaps skew-normal, with most of the weight in the less-competent zone.

saiya-jin 8y ago

For what Accenture (and probably other similar companies) delivers, for given price point, i don't get how they are still in business. Results are always super costly, and often meh or okayish at best. The variability of quality of people is simply too big and this is completely on Accenture's side. Yes some locations provide reliably good quality, but Accenture doesn't have the same standards for hiring across the globe. Seen first hand, couple of years ago batch of folks from central/east europe was hired for long term cooperation, out of 20 maybe 2-3 were worth their salt. For rest, firing them on spot would be more time-saving than baby-sitting extremely junior clueless unmotivated crowd.

1PlayerOne 8y ago

Can you tell us, in general terms, some projects that were successful engagement with Accenture?

kthejoker2 8y ago
Sure. I work in data and AI, so I'll focus on those as I know them best. We stood up and currently manage the trading platform, integration layer, data lake and analytics infrastructure, event messaging, on-demand machine learning compute provisioning and model management, API middleware, front end web app, and about 500 various service touchpoints therein of the largest oil and gas upstream finance trading desk in the world. We've developed a ton of standard demand planning and SKU-level pricing forecasting pipelines, ingest, storage, processing, model training and deployment, and consumption. In some ways that's our bread and butter as you can re-use 75-85% of the IP and focus a lot of energy on feature engineering and edge cases. We do a ton of IoT work in the mining and energy industries. Mostly (again) developing the infrastructure to ingest and process that data, building ML models to predict anomalies, production trends, workflow to connect that to SAP or some other ERP system for work orders, remote operating center dashboards, etc. We do a lot of HSE initiatives. One cool one we did recently was use some Zebra beacons and some computer vision to identify people entering dynamic "red flag" geofences (think mobile equipment, forklifts, cranes, wrecking balls, etc.) on field sites so we can instantly stop work. Anyway, I'm the first to admit Accenture's a mixed bag, but one thing I've learned since coming to work (under/for) them is large-scale delivery is very, very, very hard to 1) scope, 2) de-risk, and 3) price.