To evaluate whether a controls optimization provider is credible and unbiased for an industrial efficiency project, focus on two things:

1. Can they actually deliver measurable savings?
2. Are they incentivized to tell you the truth rather than sell you extra scope?

Here’s a practical framework.

---

1) Check technical credibility

Look for evidence they understand both controls and your process.

Ask for:

• Relevant project history in your industry and process type
• Examples of:
  • before/after performance
  • energy savings
  • uptime improvements
  • quality or throughput impacts
• The actual role they played:
  • strategy only?
  • implementation?
  • commissioning?
  • ongoing tuning?

Strong signs:

• They can explain the control philosophy in plain language
• They understand constraints like:
  • production schedules
  • safety interlocks
  • process variability
  • instrumentation quality
• They talk about tradeoffs, not just savings

Red flags:

• Vague claims like “10–30% savings guaranteed”
• No comparable references
• Overconfidence without asking detailed questions about your plant

---

2) Verify independence and incentives

A provider is more likely to be biased if they also sell:

• hardware
• software
• automation equipment
• long-term managed services tied to more spend

That doesn’t automatically disqualify them, but you should understand their incentives.

Ask:

• How are you paid?
• Is compensation tied to:
  • project size?
  • equipment sales?
  • energy savings?
  • performance results?
• Do you receive commissions from vendors?
• Do you have preferred OEMs or software partners?

Best-case model:

• Fixed-fee assessment
• Clear scope
• No hidden rebates or hardware markup
• Optional implementation priced separately
• Performance-based component only if measurement rules are agreed up front

Red flag:

• “Free audit” that turns into a sales funnel for expensive equipment

---

3) Require a measurement and verification plan

A credible provider should define how savings will be measured before the project starts.

Look for:

• A baseline period
• Normalization for:
  • production volume
  • weather
  • operating hours
  • product mix
  • downtime
• Agreed KPIs such as:
  • kWh per unit output
  • compressed air consumption
  • steam use per ton
  • cycle time
  • scrap rate
• A method aligned with recognized approaches like IPMVP if applicable

Ask:

• What exactly counts as “savings”?
• How will you avoid crediting unrelated changes?
• What happens if production changes during the test period?

If they can’t answer this clearly, credibility is weak.

---

4) Evaluate their assumptions

Good providers are conservative and transparent about assumptions.

Ask them to show:

• Which parameters are measured vs estimated
• Data quality issues
• Instrumentation limitations
• Sensitivity analysis:
  • What if the baseline shifts?
  • What if load changes?
  • What if operators override the controls?

Strong sign:

They acknowledge uncertainty and provide ranges, not single-point promises.

Red flag:

They present a highly precise ROI number without discussing uncertainty.

---

5) Look at implementation realism

A credible provider understands that controls changes can fail because of:

• poor sensor data
• operator resistance
• unstable processes
• missing maintenance
• cybersecurity restrictions
• PLC/DCS compatibility issues

Ask:

• What failure modes do you commonly see?
• How do you test logic before deployment?
• How do you roll back if performance worsens?
• How do you train operators and maintenance staff?

If they only talk about algorithms and ignore plant reality, be cautious.

---

6) Check references the right way

Don’t just ask for testimonials. Ask for references you can interview.

Questions for references:

• Did the provider hit the claimed savings?
• Were there surprises during commissioning?
• How responsive were they after go-live?
• Did they recommend changes that increased scope or cost?
• Would you use them again?

If possible, speak with:

• an operations manager
• an engineer
• someone who was involved in commissioning

That gives you a more balanced view.

---

7) Compare them against a neutral baseline

Before selecting a provider, get:

• an internal estimate if possible
• a second opinion from another independent consultant
• benchmarks from similar facilities

This helps you see whether their proposal is:

• too optimistic
• too conservative
• technically sound

A credible provider won’t be offended by comparison.

---

8) Review their reporting quality

Their proposal should be specific and auditable.

Good proposal includes:

• current-state assessment
• target systems and constraints
• proposed control strategy
• expected benefits and risks
• M&V plan
• implementation timeline
• ownership of data and IP
• assumptions and exclusions

Red flags:

• generic slide deck
• no plant-specific details
• lots of marketing language, little engineering substance

---

9) Evaluate transparency in data access

Unbiased providers are willing to let you see the data and logic.

Ask:

• Will we have access to raw data?
• Can we review setpoints, tuning changes, and alarm logic?
• Will models or code be documented?
• Who owns the data and any derived models?

If they treat the system as a black box, that’s a warning sign.

---

10) Use a simple scoring rubric

You can score providers from 1–5 on each category:

• Industry/process experience
• Controls engineering depth
• Measurement & verification rigor
• Independence of incentives
• Transparency of assumptions
• Implementation practicality
• Quality of references
• Documentation and data access

Then weight the categories based on your priorities.

A provider with a slightly lower projected savings number but strong transparency and M&V is often the safer choice.

---

Key questions to ask in a meeting

1. What similar projects have you completed?
2. How do you quantify savings?
3. What assumptions drive your ROI estimate?
4. How are you compensated?
5. What happens if the controls don’t perform as expected?
6. Can we review the raw data and logic?
7. How do you handle operator overrides and changing production conditions?
8. Can you provide references from plants like ours?

---

Short version

A credible and unbiased controls optimization provider should:

• have relevant process experience,
• explain assumptions clearly,
• define measurement and verification up front,
• disclose incentives and vendor relationships,
• offer transparent data and logic,
• and provide strong references.

If you want, I can turn this into a vendor evaluation checklist or a scorecard template you can use in procurement.