Target Data Analyst Interview Guide

Target screens for stakeholder communication harder than any other retailer you'll interview with. The interview loop includes both behavioral and culture-fit rounds alongside technical screens, and candidates consistently report that the behavioral portions carry real weight in the final decision. If your STAR stories are thin, no amount of SQL fluency will save you.

Target Data Analyst Role

You're joining a team embedded between Target's merchant organization and its data infrastructure. Depending on the sub-domain (digital/ecommerce teams lean on BigQuery, while other groups may use Hive or Spark), your daily tools will vary, but the job is the same: measure whether business initiatives are working and help category managers act on what you find. Success after year one means owning a reporting area end-to-end, where merchants come to you before making assortment or promo decisions, not after.

A Typical Week

The widget tells the story on time allocation, but what it won't convey is how interleaved the work feels. You don't get a clean "analysis day" followed by a "meetings day." A Monday morning metrics review with merchandising can spawn an ad-hoc BigQuery pull by 10:30 AM, which then feeds a Thursday readout to a VP. The packaging of insights (decks, Confluence docs, annotated dashboards) is treated as real deliverable work at Target, not busywork you squeeze in at 4 PM.

Projects & Impact Areas

Roundel, Target's media business, is one of the most active hiring areas for analysts right now, with roles focused on campaign performance measurement for advertisers across beauty, baby, and home categories. That work lives alongside more operational analytics: measuring whether a new owned-brand launch expanded the basket or cannibalized existing SKUs, or diagnosing why Drive Up conversion dipped in a specific store cluster. Target Tech also runs a less visible but genuinely interesting project area around cloud cost showback, where analysts help engineering teams understand resource allocation tradeoffs through internal dashboards.

Skills & What's Expected

Communication and business acumen are the skills candidates most consistently underweight. SQL is the daily workhorse (with BigQuery, Hive, or Spark depending on your team), and Python or R shows up for heavier wrangling. But ML and GenAI aren't part of this role. What separates hires from rejections is whether you can take a promo that lifted revenue but crushed margin and walk a non-technical VP through the tradeoff with a clear recommendation. Pipeline awareness matters at a "file a good bug with data engineering" level, not a "build the DAG" level.

Levels & Career Growth

The P2-to-P3 jump is where the role fundamentally changes: you stop executing analyses someone else scoped and start owning metric definitions, influencing what gets measured in the first place. What tends to block that promotion isn't technical skill but the inability to drive action from your analysis without a manager translating for you. Lateral moves into Roundel, supply chain analytics, or Target Tech's platform org are common and genuinely encouraged by the internal mobility culture.

Work Culture

Target requires in-office presence at Minneapolis HQ for most analytics teams, with Tuesday through Thursday being the typical cluster days. Some Roundel and digital roles are posted as remote-eligible, but this varies by quarter, so confirm during your recruiter screen. From candidate reports, the pace is steady and evenings are protected outside peak retail moments (back-to-school, holiday), though ad-hoc merchant requests can spike during those windows.

Target Data Analyst Compensation

The widget shows base, bonus, and stock by level, but the equity picture deserves a caveat. Public data on Target's RSU vesting schedule, cliff, and refresh grant cadence is essentially nonexistent for analyst roles. The stock figures above are directional estimates, and actual grants may vary by org and level. Notice that the P4 stock number is actually lower than P3 in our data, which suggests equity at Target isn't a predictable escalator the way it is at Big Tech companies. Don't model your financial decision around stock.

When negotiating, the most impactful move is confirming your level before you counter on dollars. The comp bands for P2 and P3 overlap only slightly, so getting leveled as Senior (P3) when your experience warrants it can be worth more than any base salary counter at the lower band. Beyond that, ask about sign-on bonuses. They're a one-time cost for the company, which makes them easier for a recruiter to approve than a permanent base increase. Anchor your ask with specific, quantified examples of past retail or media analytics work, since the negotiation notes indicate that scope alignment and demonstrated impact are what justify pushing toward the top of a band.

Target Data Analyst Interview Process

The process moves fast once it starts, so have your STAR stories and SQL chops ready before the recruiter call. Vague storytelling is the rejection pattern that shows up most in candidate reports, not SQL mistakes. Saying "I built a dashboard" won't cut it at Target. You need something like "I surfaced a $2M margin gap in markdown timing for owned-brand apparel, which changed the seasonal cadence."

The final loop (round 6) blends behavioral questions with SQL, product sense, and statistics, so don't treat it as a pure culture-fit conversation. Candidates who exhaust their best stories in rounds 2 and 3, then stumble when the final panel asks about stakeholder conflict or ambiguous problem scoping, put their offer at risk. Prep 6-8 distinct stories covering collaboration, delivering bad news with data, and influencing merchant partners, because Target's process will burn through a shallow bank. Practice Target-specific scenarios at datainterview.com/questions.

Target Data Analyst Interview Questions

1WITH drive_up_orders AS (
2  SELECT
3    o.order_id,
4    o.order_ts
5  FROM orders o
6  WHERE o.channel = 'DRIVE_UP'
7    AND o.status NOT IN ('CANCELLED', 'CANCELED')
8),
9item_sales AS (
10  SELECT
11    oi.order_id,
12    oi.item_id,
13    oi.dept,
14    oi.qty AS sold_qty,
15    oi.unit_price,
16    oi.line_status
17  FROM order_items oi
18  JOIN drive_up_orders o
19    ON o.order_id = oi.order_id
20  WHERE oi.line_status NOT IN ('CANCELLED', 'CANCELED')
21),
22returns_agg AS (
23  SELECT
24    r.order_id,
25    r.item_id,
26    SUM(r.return_qty) AS returned_qty
27  FROM returns r
28  GROUP BY 1, 2
29),
30net_item_sales AS (
31  SELECT
32    s.order_id,
33    s.item_id,
34    s.dept,
35    o.order_ts,
36    s.sold_qty,
37    COALESCE(r.returned_qty, 0) AS returned_qty,
38    GREATEST(s.sold_qty - COALESCE(r.returned_qty, 0), 0) AS net_qty,
39    GREATEST(s.sold_qty - COALESCE(r.returned_qty, 0), 0) * s.unit_price AS net_sales
40  FROM item_sales s
41  JOIN drive_up_orders o
42    ON o.order_id = s.order_id
43  LEFT JOIN returns_agg r
44    ON r.order_id = s.order_id
45   AND r.item_id = s.item_id
46)
47SELECT
48  DATE_TRUNC(DATE(order_ts), WEEK(MONDAY)) AS week_start_date,
49  dept,
50  SUM(net_sales) AS net_sales,
51  SUM(net_qty) AS net_units
52FROM net_item_sales
53GROUP BY 1, 2
54ORDER BY 1, 2;

1import pandas as pd
2import numpy as np
3
4
5def channel_revenue_validation(orders: pd.DataFrame, order_items: pd.DataFrame,
6                               mismatch_tol: float = 0.01) -> pd.DataFrame:
7    """Validate orders.order_total using recomputed item-level revenue and summarize by channel.
8
9    Returns a DataFrame with:
10      - channel
11      - order_count
12      - pct_orders_mismatch
13      - revenue_diff_rate
14
15    Assumptions:
16      - order_total is the authoritative total (may be null).
17      - order_items may have multiple rows per order_id.
18    """
19
20    # Defensive copy to avoid mutating caller data.
21    orders = orders.copy()
22    order_items = order_items.copy()
23
24    # Ensure numeric types; coerce bad strings to NaN.
25    for col in ["order_total"]:
26        if col in orders.columns:
27            orders[col] = pd.to_numeric(orders[col], errors="coerce")
28
29    for col in ["qty", "unit_price"]:
30        if col in order_items.columns:
31            order_items[col] = pd.to_numeric(order_items[col], errors="coerce")
32
33    # Compute line revenue and aggregate to the order grain.
34    order_items["line_revenue"] = order_items["qty"].fillna(0) * order_items["unit_price"].fillna(0)
35
36    recalc = (
37        order_items.groupby("order_id", as_index=False)
38        .agg(recalc_total=("line_revenue", "sum"))
39    )
40
41    # Join recomputed totals back to orders at order grain.
42    merged = orders.merge(recalc, on="order_id", how="left")
43
44    # Treat orders with no items as 0 recomputed total.
45    merged["recalc_total"] = merged["recalc_total"].fillna(0.0)
46
47    # Compute diff and mismatch flag.
48    merged["diff"] = merged["recalc_total"] - merged["order_total"]
49
50    # If order_total is null, you cannot assess mismatch; exclude from mismatch denominator.
51    merged["has_total"] = merged["order_total"].notna()
52    merged["is_mismatch"] = merged["has_total"] & (merged["diff"].abs() > mismatch_tol)
53
54    # Summarize per channel.
55    def revenue_diff_rate(group: pd.DataFrame) -> float:
56        denom = group.loc[group["has_total"], "order_total"].sum()
57        # If denom is 0 or missing, return NaN to avoid misleading infinities.
58        if denom == 0 or np.isnan(denom):
59            return np.nan
60        num = group.loc[group["has_total"], "diff"].sum()
61        return num / denom
62
63    summary = (
64        merged.groupby("channel", dropna=False)
65        .apply(
66            lambda g: pd.Series(
67                {
68                    "order_count": g["order_id"].nunique(),
69                    "pct_orders_mismatch": (
70                        g["is_mismatch"].sum() / g["has_total"].sum()
71                        if g["has_total"].sum() > 0
72                        else np.nan
73                    ),
74                    "revenue_diff_rate": revenue_diff_rate(g),
75                }
76            )
77        )
78        .reset_index()
79    )
80
81    return summary
82

Target's loop is structured so that a single SQL question about, say, promo lift on Tide pods can escalate into defending your metric choice to a simulated merchant partner, then into sketching the Power BI layout you'd use to present it weekly. That chain from query to recommendation to stakeholder narrative mirrors how Roundel and merchandising teams actually operate, and it's where candidates with strong technical chops but no exposure to retail workflows (comp sales, sell-through, Drive Up fulfillment rates) tend to stall. If you're only drilling query syntax without practicing how you'd frame the "so what" for a Target category lead deciding whether to expand an owned-brand, you're prepping for a different company's interview.

Practice with retail-contextualized questions at datainterview.com/questions.

How to Prepare for Target Data Analyst Interviews

Target's $15 billion sales growth plan through 2030 isn't a vague aspiration sitting in a slide deck somewhere. It's anchored to specific category bets, like the largest-ever spring beauty assortment with 60+ new brands including Supergoop and Morphe and a wellness category expansion designed to make Target the destination for affordable, personalized wellbeing.

For analysts, this means the questions you'll answer daily are tied to measurable category launches, not abstract "business growth." You might be sizing whether a new wellness subcategory is pulling incremental guests or just reshuffling existing health & beauty spend across $106.6 billion in annual revenue. Target's engineering team also publishes about internal tooling challenges like infrastructure cost showback, which signals that cost visibility and resource tradeoff analysis show up in analyst work beyond the merchandising side.

Most candidates blow their "why Target" answer by anchoring it to the shopping experience instead of the measurement problem. Three behavioral rounds means interviewers hear "I love the Target run" dozens of times a week. What separates you: pick one of those category bets (wellness expansion, beauty assortment growth, the $15B topline target) and articulate the specific analytical tension inside it. Something like, "I want to figure out whether the 60+ new beauty brands are expanding the category or cannibalizing Target's existing private-label margin, because that distinction changes how you allocate shelf space." That's a sentence only someone who studied Target's strategy could say.

Try a Real Interview Question

1WITH eligible_orders AS (
2  SELECT
3    s.campaign_id,
4    s.guest_id,
5    s.group_type,
6    s.send_date,
7    o.order_id,
8    o.order_date,
9    o.order_total,
10    ROW_NUMBER() OVER (
11      PARTITION BY s.campaign_id, s.guest_id
12      ORDER BY o.order_date, o.order_id
13    ) AS rn
14  FROM campaign_sends s
15  LEFT JOIN orders o
16    ON o.guest_id = s.guest_id
17   AND o.order_date >= s.send_date
18   AND o.order_date < DATE_ADD(s.send_date, INTERVAL 7 DAY)
19),
20first_purchase AS (
21  SELECT
22    campaign_id,
23    guest_id,
24    group_type,
25    CASE WHEN rn = 1 AND order_id IS NOT NULL THEN 1 ELSE 0 END AS converted
26  FROM eligible_orders
27  QUALIFY rn = 1 OR order_id IS NULL
28),
29agg AS (
30  SELECT
31    campaign_id,
32    group_type,
33    COUNT(*) AS sent,
34    SUM(converted) AS buyers
35  FROM first_purchase
36  GROUP BY campaign_id, group_type
37)
38SELECT
39  campaign_id,
40  ROUND(
41    SAFE_DIVIDE(
42      MAX(CASE WHEN group_type = 'treatment' THEN buyers END),
43      MAX(CASE WHEN group_type = 'treatment' THEN sent END)
44    ),
45    4
46  ) AS treatment_conversion,
47  ROUND(
48    SAFE_DIVIDE(
49      MAX(CASE WHEN group_type = 'holdout' THEN buyers END),
50      MAX(CASE WHEN group_type = 'holdout' THEN sent END)
51    ),
52    4
53  ) AS holdout_conversion,
54  ROUND(
55    SAFE_DIVIDE(
56      MAX(CASE WHEN group_type = 'treatment' THEN buyers END),
57      MAX(CASE WHEN group_type = 'treatment' THEN sent END)
58    )
59    -
60    SAFE_DIVIDE(
61      MAX(CASE WHEN group_type = 'holdout' THEN buyers END),
62      MAX(CASE WHEN group_type = 'holdout' THEN sent END)
63    ),
64    4
65  ) AS lift
66FROM agg
67GROUP BY campaign_id
68ORDER BY campaign_id;

Target's job postings for data analyst roles call out BigQuery and Power BI by name, and the Sr. Data Analyst, Digital & Ecommerce listing emphasizes transactional and store-level data work. Practice retail-flavored SQL problems (think: same-store sales comparisons, category sell-through by store format, promo vs. control period analysis) at datainterview.com/coding.

Test Your Readiness

Use datainterview.com/questions to practice defining success metrics for scenarios grounded in Target's actual initiatives, like measuring whether the wellness category expansion is driving incremental basket size or diagnosing a conversion drop tied to a specific store format change.