After the recent webinar, I was disappointed to hear that DoxCycle’s extraction accuracy might not be improved for this tax season. It was very problematic last year.
This tool could be an absolute gold mine for TaxCycle and Xero. Accurate slip extraction could save an enormous amount of time during peak season when getting the right amount of staff is so challenging, and when all CPAs get so overworked. High data extraction accuracy could hours of work every day or week for each team member
With the rise of automated prep tools in the U.S., I feel that reliable document extraction is becoming increasingly important for tax software. DoxCycle could become the Canadian version of these US prep tools.
I’m really hoping the decision can be reassessed. Even modest improvements this year would make a meaningful difference for firms like ours. We also discussed this with other firm owners, and they also really hope that this will be improved for the next tax season.
Thank you, yes, I saw this post as well. The main concern we have is the inaccuracy of the data extraction. The automated classification of documents was not causing this much of a problem. We got used to better data extraction accuracy, but this became problematic last year. So we were hoping that it would be prioritized as an improvement for this year. In a recent post, it was also mentioned that the team was planning to release an update to release DoxCycle updates to improve both the document classification and data extraction accuracy in time for the coming tax season. Therefore, we were really hoping for this improvement to happen.
Last year was unusual with CRA’s slip filing and validation changes which caused delays in data being available in AFR. They have addressed many of the issues (or so they say) for the coming tax season so we should see data getting populated in AFR earlier and more reliably this year. I agree having more reliable data extraction with DoxCycle would have been a significant help last year.
When we analyzed usage in April (peak season) only a handful of slips were actually being posted to TaxCycle (where data extraction is meaningful) about 30% of the time. Reciprocally, data extraction for ~70% of the documents processed was not required. Of those only 2 are not supported in AFR (donations and medical).
Also, data extraction relies on document classification. If the classification is incorrect that impacts the ability to properly extract the data.
For these reasons, we prioritized the classifier work ahead of the data extraction work. We, unfortunately ran out of time to complete the data extraction work in time for this tax season, but we are still working on it.
I recognize that there are a number of customers with workflows that start with data extraction and verify with AFR and there are some slips that typically are not available in AFR until later in the filing seasons (notably RRSP Contribution slips).
Our data extraction priorities are for the following slips:
T4
T5
T3
RRSP Contributions
Medical
Donations
Happy to get feedback on the prioritized slips and/or add you to our beta tester pool for DoxCycle.