Visualize Your Job Search With a Sankey Chart

A job search can feel busy without feeling clear. You might have sent dozens of applications, had a few promising calls, and waited on several employers at once. A spreadsheet full of dates and notes still makes the whole process hard to see. A Sankey chart turns those records into paths, with wider ribbons showing how many applications moved between each stage.

Treat the chart as a way to decide what to review next. It turns a vague impression into a useful picture. You can see whether referrals are reaching interviews, whether most applications are ending before a screen, and where late-stage opportunities tend to stop.

What a job-search Sankey chart shows

A Sankey diagram connects stages with flows. In a job search, a source can be an application channel or hiring stage, a destination can be the next stage or final outcome, and the amount is the number of applications that made that transition. The chart preserves branches and merges that a single conversion percentage can hide.

For example, direct applications and referrals can both lead to a recruiter screen. Some screens move to a hiring-manager interview, while others close. Later interviews can split into offers and no-offer outcomes. A Sankey chart keeps those routes visible in one view.

This is different from a standard funnel. A funnel assumes one main sequence and usually shows how a total shrinks at each step. A Sankey chart is better when applications enter through different channels, skip stages, return different outcomes, or need to be compared as paths rather than as one stack of percentages.

Start with an application tracker you trust

Build the chart from records, not memory. At minimum, track the employer, role, application date, source or channel, current stage, last meaningful update, and final outcome. CareerOneStop's job-application guidance recommends recording the positions you applied for and employers you contacted. That record also helps with follow-ups and prevents duplicate applications.

A simple spreadsheet or a board such as the utilkit Kanban Board can hold the live details. Keep one row or card per role. Use the same small set of stage names every time so “phone screen,” “recruiter call,” and “intro call” do not become three labels for the same event.

A practical stage list might be:

• Saved or identified
• Applied
• Recruiter screen
• Hiring-manager interview
• Final interview
• Offer
• Closed or withdrawn

Do not force every employer into every stage. If an application moves directly to a hiring-manager interview, record that actual transition. If you withdraw, use a separate outcome rather than grouping it with an employer rejection. The distinction matters when you later ask why applications left the process.

Choose a clear time window

Choose a time period before you count the paths. You might use applications submitted during the last 60 or 90 days, or one complete search from its start through an accepted offer. Mixing a recent week with applications that have had months to progress makes the older applications look more successful than the newer ones.

Applications still waiting need an honest label. Add an “In progress” destination, or leave the newest applications out of an outcome comparison until enough time has passed. Silence is not automatically a rejection, and an open application has not yet followed a final path.

Also decide what one unit means. For this chart, use application counts throughout. Do not mix counts, percentages, hours spent, and expected salary in the same diagram because ribbon widths would no longer be comparable.

Turn application records into flow rows

For each application, list every transition it completed. Then group identical From/To pairs and count them. A spreadsheet can do this for you, but you can count a small search by hand. The result needs only three columns: From, To, and Amount.

Here is a consistent example for 40 applications:

Paste these columns into the utilkit Sankey Chart Maker. The editor accepts From, To, and Amount headers as well as Source, Target, and Value. Keep labels consistent and amounts positive. The chart rejects circular paths, so treat a later reapplication as a separate application instead of drawing an arrow back to an earlier stage.

Read the ribbons carefully

The widest early ribbon in the example runs from direct applications to roles that closed before a screen. Start by reviewing your targeting, resume alignment, role level, and application quality. The chart cannot prove that one resume sentence caused the result. Hiring decisions also reflect competition, timing, internal candidates, changed budgets, and information you cannot see.

The referral path is smaller but reaches recruiter screens more often in this sample: 4 of 10 referrals compared with 6 of 30 direct applications. Use the Percentage Calculator when you need exact conversion rates. A Sankey ribbon shows volume, but a small sample does not prove a lasting pattern.

At later stages, three final interviews produced one offer. That might suggest the overall approach is working while final-stage preparation could improve. Before changing anything, read the notes behind those interviews. Three interviews are too few for a firm conclusion, and the roles may have required different skills.

Questions worth asking

Use the chart to answer practical questions:

• Which channels produce interviews, not just applications?
• Are you applying mostly to roles that match your experience and location needs?
• Where do applications most often leave the process?
• Do referred and direct applications behave differently?
• Are enough recent applications still open to distort the apparent rejection rate?
• Did your results change after you updated your resume, portfolio, target roles, or interview preparation?

Answer one question at a time. If you change your resume, target role, and application channel in the same week, the next chart cannot tell you which change mattered. Record meaningful changes and compare one period with another carefully.

Common mistakes

Counting current status instead of transitions

If six applications reached a screen and three later reached a final interview, both transitions belong in the data. Recording only each application's current stage loses the path between stages and prevents the flows from balancing.

Combining very different searches

Entry-level design roles, senior engineering roles, contract work, and internal transfers may have very different processes. Split them into separate charts or add clear source categories. A single attractive diagram is not useful if its labels hide important differences.

Treating every rejection as feedback

A flow shows what happened, not why. Use it to choose where to investigate, then read recruiter feedback, interview notes, job requirements, and your own constraints. Do not turn a small chart into a verdict about your ability or future prospects.

Ignoring the starting total

Five interviews can be encouraging or disappointing depending on whether they came from 10 applications or 100. Keep the starting total visible, compare counts with rates, and add the time period when you save or present the chart.

Protect private job-search information

Remove identifying details before you share the chart. Use labels such as “Referral,” “Direct application,” and “Final interview” instead of employer names, recruiter names, email addresses, or confidential compensation details. The Sankey maker processes data in your browser, but a share link with “Include my inputs” selected contains the chart data in the URL. Anyone who receives that URL can read those inputs, and the address may remain in browser history or other logs.

Manual node positions are not included in the shared link. People who open it get the flow rows, colors, and chart settings, and the diagram calculates a fresh layout. Export a PNG or SVG when the exact dragged arrangement matters.

Make the chart part of a weekly review

Update the tracker as applications move, but rebuild the chart on a steady schedule rather than after every rejection. A weekly or biweekly review is often enough to catch missing follow-ups, applications with no recent activity, and stage changes without turning the search into constant measurement.

Pair the chart with information about the wider job market instead of relying only on your response rate. The U.S. Bureau of Labor Statistics' job-search guide points to occupation requirements, outlook, and wage data that can help you judge whether a target fits your experience and location. CareerOneStop's broader job-search resources cover planning, networking, applications, interviews, and negotiation.

A useful job-search Sankey chart does not need dozens of nodes. Start with the smallest set of stages that answers your question, keep the counts easy to check, and revise your approach only when the chart and your notes point in the same direction.