Q 777. During the Analyze phase, how do you dig into what’s really breaking a process—like a retail chain seeing sales drop because of stockouts, only to find it’s tied to poor forecasting, not delivery delays? What’s your go-to way to separate the real causes from the noise or symptoms, and how do you prevent mistakes like chasing the wrong cause?

 

Note for website visitors -

• This platform hosts two weekly questions, one on Monday and the other on Thursday. 
• All previous questions can be found here: https://www.benchmarksixsigma.com/forum/lean-six-sigma-business-excellence-questions/. 
• To participate in the current question, please visit the forum homepage at https://www.benchmarksixsigma.com/forum/.
• The question will be open until Monday or Thursday at 5 PM Indian Standard Time, depending on the launch day. 
• Responses will not be visible until they are reviewed, and only non-plagiarised answers with less than 5-10% plagiarism will be considered for winner selection. 
• If you are unsure about plagiarism, please check your answer using a plagiarism checker tool such as https://smallseotools.com/plagiarism-checker/ before submitting. 
• All correct answers shall be published, and the top-rated answer will be displayed first. The author will receive an honourable mention in our Business Excellence dictionary at https://www.benchmarksixsigma.com/forum/business-excellence-dictionary-glossary/ along with the related term.

• Some people seem to be using AI platforms to find forum answers. This is a risky approach as AI responses are error-prone because our questions are application-oriented (they are never straightforward). Have a look at this funny example - https://www.benchmarksixsigma.com/forum/topic/39458-using-ai-to-respond-to-forum-questions/

• We also use an AI content detector at https://quillbot.com/ai-content-detector. Only answers with less than 45-50% AI-generated content will be considered for winner selection.

Recommendations for finding the actual root cause:

• Perform root cause analysis with 5 whys 
• Plot the full process with swim lanes for identifying the bottlenecks and inefficiencies 
• Interview the people 
• Look out for patterns in the data 

 

Recommendations for avoid chasing the wrong cause:

• Cross check the data collected 
• Validate the findings from different angles
• Discuss with people part of the process to correlate their observations

Digging into What's Really Breaking a Content Moderation Process (Analyze Phase)

To separate real causes from noise or symptoms, in the Analyse phase, a structured data driven approach is required. RCA techniques, stratification and validation methodologies would help identify the right root cause. 

 

Data Management: 

Ensure appropriate & accurate data is collected is the first time to identify actual root causes. 

1. Process Mapping & Identification of data points: Create a detailed process map of the Workflow, identifying each step from input to output and data storage. For each step identify all the possible data points to be collected. Consolidate identified data points and proceed to collect data

 

2. Data collection & analysis: Collect the data and conduct multiple analysis like Trend analysis, Pareto, SPC & Cycle time analysis on the quantitative data. In addition, depending on the problem statement, we may need to collect & analyse qualitative data as well. 

 

5 Why analysis, Fishbone analysis, FMEA and Cause & effect diagrams are some of the statistical tools that can be used to identify the reason for the breakages in the process. 

 

Separating Real Causes from Noise/Symptoms

• validation of the data for any hypothesized cause. If a cause is identified, there must be a quantifiable evidence that the cause directly impacts the problem. 
• Is the problem consistently caused by the problem. If the cause only occurs sporadically, then the identified cause May be a contributing factor to another root cause.
• Conduct a PoC to see if elimination of the data backed up cause impacts the problem. 
• Assess the potential impact that is expected by addressing the cause.
• On the basis of the impact, prioritise the cause with the highest potential impact. 

Preventing Mistakes (Chasing the Wrong Cause)

• Concluding the causes only with data validation to avoid chasing the wrong cause. 
• Reduce/eliminate any assumptions with regard to identifying the cause
• Utilise multiple data points to triangulate the cause 
• Involve diverse stakeholders during data analysis to uncover different perspectives and challenge assumptions

 

Vidhya R

 

 

Start with Process Data + Visual Tools

Use the data collected during the Measure phase and present it visually with:

• Pareto charts
• Histogram
• Control charts
• Scatter plots

Example: If a significant number of patient delays occur in a specific department or shift, the Pareto chart will highlight this.

 

Visit the Gemba (the actual location)
Process issues are infrequently completely apparent from a conference room.

Root Cause Analysis
• 5 Why Analysis - Continue asking “why” until you identify an actionable cause
• Fishbone (Ishikawa) Diagram - Organize causes into categories: People, Process, Equipment, Environment, Materials, and Management.

In the Analyze phase, it's important to get vevery team member involve. Brainstorming / brainswritting to identify potentiel causes. Never assume always focus and based on facts.  Group possible causes and confirm root causes with data some time additional data need to be collected for confirmation. most of the time Analyse and Improve need several iteration till the root cause can be identify.

In the sales case simple 5 why could show that stockout is a consequence and not a root cause

It's where AI could help getting real data more releable than historical.

I would approach this problem using the DMAIC approach. Understanding the different stores, products and timeline.

 

1. With these, I would need to validate the issue and understand the impact using data,

Example is sales down by 10% in stores with 16% stockouts. Categorize the data by stores, products and time. 

 

2.  After completing process 1. This would lead me to ask the right questions which is my root cause analysis. 

 Using the 5 whys approach,

• Why are sales dropping? Stockouts.

• Why stockouts? Inventory not replenished.

• Why not replenished? Forecast underestimated demand.

• Why was demand underestimated? Forecast model doesn’t account for promotions.

• Why? Forecasting model lacks integration with marketing calendar.

I would use Fish bone diagram to categorize potiential causes across. This will help me explore all possible causes before diving into a specific cause. Example of potential cause in this case could be:

Forecasting methods, machines, environment, seasonality. 

 

3. Next step is to run a pareto analysis to find the impacted categories. Additonally, I will run a correlation study to validate the relationship between forecast accuracy and stockouts. I can track trend analysis which can show if stockouts is due to seasonality of if this is event based.

 

4. While going through these process, I will ensure my data collection process is unbiased or there are no errors in data collection process. Additionally, I can collects anecdotes to track correlation between the anecdotes and the data. 

 

5. Lastly, I will prioritize root cause based on correlation with the problem and how they can be controller. 

The 'Analyze Phase' is the third stage of a DMAIC project. The Analysis phase is purely about precision and clarity, taking inputs from the previous 'Measure' phase and generating outputs for the next 'Improve' phase. I resist the urge to dive right into averages or superficial statistics when I begin the Analyze phase. Instead, I look for patterns and the real core causes of process variance using Lean Six Sigma methods. Purposeful analysis is fuelled by a targeted problem statement that is founded on VOC and process mapping. There are many visual aids, such as box plots, histograms, and run charts can be used to highlight odd patterns and changes. To find the areas with the highest concentration of issues, regular process gaps, I stratify data by time, provider, or visit type. Pareto charts rank the most important problems; on the other hand, control charts separate normal fluctuation from unique reasons. To delve deeply and refrain from making snap judgments, I recommend the Fishbone diagrams and the Five Whys. Data is validated through triangulation—Gemba walks, manual logs, and system data comparisons. The Analyze phase is where I turn raw data into insight, targeting variation, validating root causes, and aligning improvement with what matters most to the customer and the business.

In the Analyze phase we need to do a Fish bone analysis to identify all the causes that are impacting the outcome. By using the Pareto analysis we can identify what are the major causes that have an impact on the output. We can then do a Why - Why analysis for each cause to identify the root causes.

 

Basis the data collected in the earlier phases of the DMAIC approach, we will use hypothesis testing on the high impacting root causes. This will confirm statistically if the root cause we have identified is indeed causing the  impact.

 

The Why Why analysis to identify the root causes and a hypothesis testing usually confirm what kind of relationship exists between the identified Xs and final output Y. Taking a data driven approach and accepting the statistical outcome will always confirm what are the real causes and what is just a symptom.

Below are some of key steps which can be taken to ensure real causes and being identified and pursued in the Analyze phase –

 

1.)    Use of structured data – It is very important to use a structured, data driven approach and to avoid relying on anecdotes.

2.)    Ensure the problem statement is specific and quantifiable.

3.)    Ensure proper data stratification.

4.)    Follow 5 Why effectively – Do not stop at the first why, keep digging until a proper root cause is revealed.  

 

In the Analyze phase, to find out what's really breaking the process instead of fixing the noise from the process,

1. List down the entire process from start to end; it will help visualize the data in each section where it goes wrong in the process.

In the given example, the process mapping is as follows:

• Sales Forecast
• Procurement
• Production/Order from supplier
• Items in Transit
• Items in Warehouse Stock

2. Root cause analysis techniques must be implemented, such as the 5 Whys technique and fishbone diagram, to find out the cause of the problem.

By doing the 5 Whys technique, root cause analysis as, 

 

Problem statement: Sales are dropping because of stockouts.

1. Why? Not enough products on the shelf

2. Why? There is no item stock in the warehouse and other distribution centers.

3. Why? The procurement team did not make enough orders for this item.

4. Why? The order by the procurement team is based on the sales forecast, which is inaccurate.

5. Why? The forecast model by the team did not consider the seasonal promotions and order demand.

Here the problem is not the delayed delivery; it is because of poor sales forecast planning not considering the seasonal days it follows the previous sales historical data.

 

How to prevent the mistake of chasing the wrong cause?

1. Look into the data; check the actual numbers, not by guess.

2. Involve all the stakeholders in the process and discuss and analyze with them before concluding the solution to the problem.

3. Focus on the process, not the people.

Like any other project- during Analyze phase, we must first define the problem well, and thereafter break the process into steps to determine bottlenecks and root causes.

 

Approach- 

1. Prepare the current state process map.

2. Do a few brainstorming sessions to identify potential causes.(Ishikawa diagram and 5Whys)

3. Map the potential causes to steps/stages in the process map

4. For shortlisted potential causes do few hypothesis testing to validate the data/causes.

 

This approach will help us separate real causes from noise.

So by using structured root cause analysis and hypothesis testing techniques, we can validate the findings and propose targeted corrective strategies.

In addition, we need to start by analyzing different sets of data to understand trends, timelines, forecast accuracy etc.

Data sources may include:

Sales Data (12 months): Seeking trends to indicate a mismatch between high-demand periods and stock availability leading to Lost Sales impact.

Inventory Records: For stock-out frequency of high-volume SKUs across the above period.

Forecast Accuracy Reports: To check for variance from actual demand.

Delivery Timeline Logs: For Supplier delivery windows. On-time delivery performance >95%, suggesting minimal impact from logistics delays.

Key Insights identified :

Forecast Error Correlation: High forecast inaccuracies coincide with stockout events, especially for promotions and new product launches.

Replenishment Lag: Time between forecast input and stock arrival often spans 10–14 days, reducing responsiveness.

Demand Volatility Ignored: Forecasts do not factor in localized demand surges, social trends, or weather-related events.

Inventory Turnover Analysis: Low turnover rates in some categories suggest misallocation of inventory resources.

 

Analysis Techniques Used

1. Process Mapping (SIPOC Analysis)

2. Root Cause Analysis - 5 Whys Technique

Problem: Products are frequently out of stock

3. Fishbone Diagram (Cause & Effect Analysis)

4. Hypothesis Testing

Test 2 primary hypotheses using statistical analysis:

Hypothesis 1: Delivery delays cause stockouts. We found out that on-time delivery has remained stable at > 95%, so we can confidently disprove this hypothesis. It's noise, not the cause.

Hypothesis 2: Poor forecasting drives stockouts. By mapping “Forecast accuracy %” against “stockout incidents” we can find out the correlation between them. With a significant p-value we have statistical proof that this is the primary driver.

How to Prevent Chasing the Wrong Cause -

 

Define the Problem with Data - Begin with a problem statement from the Measure phase (e.g., "From Q2 to Q4, stockouts on A-list items increased by 40%, contributing to an estimated $1.2M in lost sales."). This is our ultimate benchmark. If an identified "cause" doesn't statistically impact this metric, it's not the right one!

Gemba walk – Speaking with the store managers, the inventory planners, and the logistics coordinators is important as they have qualitative insights that can help with data analysis.

Use a combination of Tools: Using a combination of tools like the Fishbone Diagram provides the structure, the Regression Analysis provides the statistical proof, and the 5 Whys provide the deep dive.

Using the symptom identify the root cause, in the following example sales drop is the symptom due frequent stockout. Stockout happening across all locations, products and check the historical data points. Once i have the answer using fishbone or 5 why analysis identify the root cause why sales drops --> Why products not available --> Why inventory not replaced on time --> Root cause poor forcast and next Fishbone analysis Is the challenge because of People, process, technology. This helps to identify the root cause visually to correlate with the symptom. perform hypotheses test with data. Chasing the symptom instead of root cause leads no permanent solution! 

In DMAIC  analyse phase of Six sigma data that has been gather in a previous phase (for eg measure phase) . It can help to find out places where improvements are needed.

 

The analyse phase will concentrate on finding out why a certain issue is occurring. It uses various statistical tools to analyse data and find reasons for the problems.Since it focuses on finding the underlying cause of a problem it can be used to separate the real cause from the noise .Careful sampling is needed for the success of the analyse phase.

 

To separate the real causes from the noise so as to prevent mistakes we can combine critical thinking, analysis and observation. Investigation of potential causes and looking into the patterns and isolating the variables can lead to better analysis.

Repeatedly asking the WHY questions can help to drill down the problem

 

Also quickly jumping to conclusions and ignoring the obvious or ignoring the big picture can lead to pitfalls. Advanced tools like ANOVA , Regression and hyposthesis tesing can be used

To get to the heart of the issue, it’s essential to go beyond surface-level symptoms.

Here are a few practical ways to separate the noise from the real root causes:Its really critical to understand the root cause analysis and following are the ways to separate noise from real causes.
1.Start probing more with the “5 Whys” approach. This classic technique wont stop until getting to the core problem
2. Look at from multiple data points and right range - Look at the issue from multiple data points: Operational data (e.g., delivery logs), Customer data (e.g., complaints, churn rates), Market data (e.g., competitor moves, seasonal trends).
3. Create a visual represenation using Current reality tree - Tools like CRT, SIPOC diagrams or value stream mapping help visualize where handoffs or delays occur. 

In order to prevent chasing the wrong cause, ensure you are following a structured problem solving framework like DMAIC, reviewing KPIs to catch early signals of issues and validating the hypothesis with right data points.

Sargun has provided the best answer to this question. Well Done!!