Decide who gets a call before you offer one. Describe your service and Taskade Genesis builds a branching qualification map: the questions that matter, what each answer should lead to, and where the path ends in a booked call, a resource, or a warm no.
The Consultancy Booking Form is a live Taskade Genesis app on this page. Answer its questions, see how a real qualifying form behaves, then press "Use this app" to clone it in about ten seconds and open it out into the branching discovery funnel described here.
Most discovery call funnels are built backwards. Someone writes six questions, adds them to a form, and only discovers months later that two of the questions decide everything and four are decoration. Mapping the branches first shows you which answers actually change the outcome, and a map is far easier to argue about with a colleague than a live form.
What the build gives you: a visual tree of the qualification path, the question wording at each node, the routing rule for each answer, an outcome at every leaf so nothing dead-ends, the message that goes with each outcome, and a note of which question is doing the real work.
The map renders across the layouts you need:
- Mind Map view for the branching path, which is the view that makes gaps obvious
- Board view to group nodes by outcome: Book, Nurture, Refer, Decline
- Table view for question wording, answer options, and routing side by side
- Plus List, Calendar, Gantt, and Org Chart across the rest of the 7 project views
The map becomes the build. Once the branches are agreed, the same app carries them into a real form using form logic, with a branch action applying the routing and a filter sending high-fit enquiries straight to a person while everyone else receives the right resource. Across 100+ bidirectional integrations, web forms and Gmail pull enquiries in while email and Slack push routing out. Automations run on paid plans; the Free plan includes 10 flow runs in total.
The AI agents in the app carry 34 built-in tools including web search, persistent memory, and multi-agent collaboration. The Funnel agent proposes the question set from a description of your ideal client, spots branches with no outcome attached, and rewrites questions that are really two questions wearing one coat.
The discipline that makes this work is insisting every leaf has an outcome. A funnel with a dead end sends someone who took the trouble to answer your questions into silence, and silence is the most expensive response in the whole process. A referral, a guide, or an honest "not the right fit, here is who is" costs nothing and is remembered.
Question order matters more than question count. Ask the disqualifying question early, before the effort is sunk, and the person who was never going to be a fit leaves in twenty seconds rather than three minutes. That is kinder to them and produces cleaner data for you.
Clone the app, invite whoever owns your pipeline, and the map is yours to argue over and change. That is the point of drawing it before you build it.
Learn form logic in Learn, branching in Learn, and app context in Learn. Browse live builds in the Community Gallery. Pair this with the consultation intake queue, which is where the qualified enquiries land, and the request and confirm form that turns a qualified enquiry into a booking.
Frequently Asked Questions
Why map it before building the form?
Because a map shows you the branches with no outcome, the questions that never change the routing, and the places where a good-fit enquiry can fall through. Finding those on a diagram takes an hour. Finding them in production takes a quarter.
How many questions should a discovery funnel have?
Usually fewer than you think. If an answer does not change what happens next, it is not a qualifying question, it is research, and research belongs on the call rather than in front of it.
What should happen to people who do not qualify?
Something useful. A guide, a referral, or an honest explanation. Sending nothing is the worst option available, and it is what happens by default when a branch has no outcome attached.
Can the map become the actual form?
Yes. The routing you agree on the map is implemented with form logic and a branch action in the same app, so the design and the running funnel do not drift apart.
How do I know which question is doing the work?
Record the routing outcome against each answer and, after a few dozen enquiries, one or two questions will account for nearly all the decisions. The rest can usually be cut, which improves completion rates.
Should the funnel book the call directly?
Offer the booking to qualified enquiries only, and route everyone else elsewhere. That is the whole point of qualifying first, and it is what stops your calendar filling with calls that were never going anywhere.
Can a team work on this together?
Yes. Invite colleagues to the workspace and edit the map together. Disagreements about who is a good fit are much easier to settle in front of a diagram than in an argument about a form field.
Does this work for inbound and outbound?
Yes. Inbound enquiries walk the branches themselves. For outbound, the same tree becomes the qualification script, which keeps the standard identical whichever direction the conversation started from.
