The Future of Credentially Webinar: Q+A

We had a brilliant response to our Future of Credentially session — and the questions kept coming. Here are all your questions, answered.

"Will these tasks in agent activity and needs attention show up for all admins or only the admins that are assigned to those staff members - is there an option to change this?"

Great question and one we've had before. Yes, tasks will be visible to all admins so that if someone is off or unavailable, a colleague can pick it up and move things forward. We'll also support the ability to filter by assignment so you can focus on your own workload when needed. The goal is to keep things collaborative while still giving you a clear view of what's on your plate.
‍

"Having various systems for different purposes always causes problems and duplication. Can we fully integrate Credentially with our CRM to avoid these issues?"

Completely agree, and this is exactly why we've prioritised our API work this quarter. We already integrate with a number of CRMs and we're continuing to expand that. The goal is to support the full end-to-end workflow so data flows between your CRM, Credentially and any other systems you use without anyone having to re-enter information. If you'd like to discuss your specific CRM setup, your CSM can arrange a call with us to look at what's possible today and what's coming.
‍

"Do you have plans on the roadmap to make an AI companion available to candidates going through onboarding to reduce queries to us as an organisation and the Credentially support team?"

Yes, this is on our roadmap. We're developing what we call an AI Compliance Companion that will be able to guide candidates through the onboarding process. It'll be able to answer their questions about what's needed and why, based on your organisation's specific policies and requirements. The first version will work via email, and over time we'll expand to other channels. The idea is that candidates get immediate, accurate help without needing to contact your team or ours for routine questions.
‍

"How does this AI learn our specific policies/requirements? What can we do if we know the AI has provided incorrect information?"

The AI learns your policies through a knowledge base that you control. You upload your policies, procedures, SOPs and FAQs, and the AI references those when making decisions or writing content. It always cites its sources so you can see exactly where it pulled information from. If the AI gets something wrong, there are a couple of safeguards. First, for important decisions the system uses human-in-the-loop, meaning it will create a task for someone on your team to review before an action is taken. Second, you can update the knowledge base at any time, and the AI will immediately reflect those changes. We're building this with full transparency so you always understand why the AI made a particular decision.
‍

"Can the task list also track tasks that aren't done in system - ie send email to IT for IT set up?"

Yes. The agent framework is built with a set of tools, and some of those tools make API calls to external systems. So if you need to trigger an action outside of Credentially, like sending a request to IT, the system can support that. We also see tasks as a general-purpose tool within the platform, so you'll be able to create manual tasks for activities that happen outside the system and track their completion alongside everything else.
‍

"If a candidate completes their onboarding will they automatically move to the next stage in placements or do you have to manually move them? The same for showing as compliant, will they automatically move stage?"

That's the plan. When a candidate meets all the compliance requirements for a placement, the system will be able to automatically update their status and move them through the pipeline. We want to remove as much manual stage-management as possible. The rules for what counts as 'compliant' are defined by you through the compliance packages, so once those conditions are met the system can act on it. We'll share more detail on this as we get closer to the placements release.
‍

"Is there any plans to link the DBS update service check to the DBS document upload so a live update service check can link to the validity/expiry date of the document requirement?"

Yes, DBS improvements are on the roadmap. We recognise that the current setup treats new applications and update service checks as separate things when they're really part of the same compliance picture. We're looking at linking DBS update service checks directly to the document requirement so that expiry dates and check status are connected. This includes better handling of different expiry policies, for example CQC's three-year rule for new applications. More detail to come as we scope this out properly.
‍

"Work History - Will the AI capture if the gap was time out for additional education?"

Yes. As we showed in the demo, the work history module supports configurable activity types. That includes education and training alongside employment, career breaks, travel and other activities. When the AI extracts information from a CV, it'll pick up education periods. And if there's a gap, the candidate can specify the reason, including further education. The opportunity is to capture a complete picture of what someone was doing, not just their employment history.
‍

"If you request 5 years refs from work history but someone has worked in the same place for 10 years, will you be able to tell it to request 2 references from the same place?"

Good scenario. The referencing rules will be configurable per organisation, so you'll be able to define things like minimum number of references, which types of roles require references and how far back you need to go. For a situation where someone has been with one employer for a long period, the system will support requesting multiple references from the same organisation. We know this is a common scenario and it's something we're building into the configuration.
‍

"Is the dashboard and all information visible to admins with restricted permissions? Eg, if they have restrictions on the documents able to view but have admin permissions in other areas?"

Existing permission controls will carry through to any new features. So if an admin has restricted access to certain document types or candidate groups, those same restrictions will apply to what they see on dashboards, in tasks and across the AI features. We're not building these new capabilities outside of the permission model. Your existing setup will be respected.
‍

"Is there an audit trail for all emails sent to a candidate? Can all admins see the emails sent to candidates? Can a candidate reply directly? If so, where does the reply go?"

Yes to all of those. Every communication sent from the platform, whether by a person or by an AI agent, is logged with full visibility. Admins can see the complete history of messages sent to a candidate. Candidates can reply directly, and those replies come back into the platform. As we showed in the demo, the system can then process those incoming messages, update information where appropriate and escalate to a human when needed. Everything is tracked in the candidate's timeline so you've got a full audit trail.
‍

"For the certificate checking, is this something that can automatically be done and the admin just goes in and checks, or does the agent have initiate the check?"

Both. When a document is uploaded, the system will automatically classify it, extract information and run the verification process. The agent handles the end-to-end flow, including visiting third-party websites to cross-reference details, as we showed with the BLS certificate example. If everything checks out, it can approve the document. If something doesn't match or confidence is low, it creates a task for a human to review. So you're not having to initiate the check manually, but you'll always have the option to step in and review.
‍

"Do you anticipate introducing the option to trigger a compliance package at a certain point of onboarding. eg once a candidate has started or been signed off to then automatically trigger another compliance package"

Yes, this is something we're exploring as part of the placements work. The idea is that compliance isn't always a one-time thing. Some requirements might only apply once a candidate is placed, or you might need to trigger a different set of checks at a certain milestone. We want to support that kind of conditional triggering so compliance packages can adapt to where the candidate is in the process rather than treating everything as upfront onboarding.