Mason

Robotics hardware company

How we helped
a robotics company
find its next customers.

Read the case
ROBOTIC HANDS & ADVANCED GRIPPERSFrom buyer research to relevant outreach

ABOUT THE COMPANY

A better grip.
A different kind of buyer.

Our client develops adaptive robotic hands. Its next customers are robot makers, integrators and teams building machines that need to handle real objects.

THE BRIEF

Find the programmes
making their next hardware decision.

A company building robots is not automatically buying an external hand. We studied comparable products, their customer relationships and the changes that make integration worth discussing.

Buyer researchAccount researchRelevant outreach
PHASE01

Understand who buys.

Compare the actual hand, the application and the customer. Follow the evidence behind each relationship.

01 / THE STARTING POINT

Start with the hand. Then the application.

We mapped the component our client sells and the work it needs to do: adapt to real objects, mount on a robot and fit the application’s control requirements. Whole robots and simple grippers were different comparisons.

The robotic handA precise application to study
THE STARTING POINTIN ACTION
THE OFFER

We mapped the component our client sells and the work it needs to do: adapt to real objects, mount on a robot and fit the application’s control requirements. Whole robots and simple grippers were different comparisons.

The product scope+

The public product category is adaptive robotic hands and advanced grippers. Unpublished client product names, price, roadmap and private specifications are not included.

02 / THE COMPETITOR UNIVERSE

Find the makers solving the same problem.

We ran 252 searches and read 225 results pages. That produced 179 makers and adjacent companies. Ninety-eight belonged in the comparison; resellers, whole robots and unrelated products were set aside.

179 market names98 relevant makers
THE COMPETITOR UNIVERSEIN ACTION
THE WIDER MARKET

We ran 252 searches and read 225 results pages. That produced 179 makers and adjacent companies. Ninety-eight belonged in the comparison; resellers, whole robots and unrelated products were set aside.

The search scope+

252 queries; 225 result pages read; 179 market names; 98 kept. The comparison excluded resellers, whole robots with captive hands, simple grippers, prosthetics-only makers and research-only projects.

03 / THE CLOSEST COMPARISON

Choose the closest twenty.

We scored the closest product from each maker on product, application, customers, ecosystem and offer. Product carried 45% of the score. Twenty suppliers became our route to real customer evidence.

Five weighted questions20 closest suppliers
THE CLOSEST COMPARISONIN ACTION
THE COMPETITOR SHORTLIST

We scored the closest product from each maker on product, application, customers, ecosystem and offer. Product carried 45% of the score. Twenty suppliers became our route to real customer evidence.

Scores and sources+

Product 45%, Application 20%, Clientele 15%, Ecosystem 10%, Offer 10%. The closest offer is compared, not every product sold by the company.

  • ChangingTek ↗ · Scored 75 of 100 on its closest product: Shuqiao (CTDH5F80).
  • Tesollo ↗ · Scored 74 of 100 on its closest product: DG-5F-S (Delto Gripper 5-Finger, Short/light version).
  • PSYONIC ↗ · Scored 72 of 100 on its closest product: Ability Hand.
  • Robotiq ↗ · Scored 68 of 100 on its closest product: 3-Finger Adaptive Gripper.
  • Sharpa ↗ · Scored 67 of 100 on its closest product: Sharpa Wave.
  • Linkerbot ↗ · Scored 66 of 100 on its closest product: Linker Hand O6.
  • Inspire Robots ↗ · Scored 62 of 100 on its closest product: RH56F1.
  • qb Robotics ↗ · Scored 62 of 100 on its closest product: qb SoftHand Industry.
  • ROBOTIS ↗ · Scored 59 of 100 on its closest product: HX5-D20 (HX5-D20-MRT / -MLT).
  • AGILINK ↗ · Scored 58 of 100 on its closest product: OmniHand 3 (Standard).
  • BrainCo ↗ · Scored 58 of 100 on its closest product: BrainCo Revo 2 Pro (five-finger dexterous hand, EtherCAT/industrial variant).
  • Alt-Bionics ↗ · Scored 56 of 100 on its closest product: Surge (robotics hand).
  • DH Robotics ↗ · Scored 56 of 100 on its closest product: DH-5 / DH-5-6 dexterous hand.
  • Leadshine ↗ · Scored 56 of 100 on its closest product: DH116 (11 DoF, 6 active actuators, 545 g, worm-gear drive).
  • OnRobot ↗ · Scored 56 of 100 on its closest product: 3FG25 heavy-duty 3-finger electric gripper.
  • Apicoo Robotics ↗ · Scored 55 of 100 on its closest product: SusGrip-3F (3-finger concentric adaptive electric gripper).
  • SCHUNK ↗ · Scored 54 of 100 on its closest product: SVH 5-finger servo-electric gripping hand (SCHUNK SE & Co. KG).
  • Aeon Robotics ↗ · Scored 53 of 100 on its closest product: HandEffector.
  • Ruiyan ↗ · Scored 53 of 100 on its closest product: RY-H1.
  • COVVI Robotics ↗ · Scored 53 of 100 on its closest product: COVVI Robotic Hand.

04 / THE CUSTOMER EVIDENCE

Find where their hardware is being used.

We followed supplier pages, buyer technical accounts, integration reports and public records. The twenty suppliers named 519 customers and users. We kept the application and relationship type with each name.

20 suppliers519 customer and user records
THE CUSTOMER EVIDENCEIN ACTION
THEIR CUSTOMERS

We followed supplier pages, buyer technical accounts, integration reports and public records. The twenty suppliers named 519 customers and users. We kept the application and relationship type with each name.

The evidence routes+

519 public customer/user records → 159 dated companies → 39 that fit the customer profile → 21 studied relationships. Records include customers, research users, integrations, demonstrations and commitments; those are not interchangeable.

05 / THE RESEARCH COHORT

Choose relationships we can actually study.

Of the public records, 159 companies had dates and 39 also fit the target customer profile. We selected 21 relationships. A purchase, an integration, a letter of intent and a demonstration stayed distinct.

39 companies fit21 relationships studied
THE RESEARCH COHORTIN ACTION
THE STUDY SET

Of the public records, 159 companies had dates and 39 also fit the target customer profile. We selected 21 relationships. A purchase, an integration, a letter of intent and a demonstration stayed distinct.

The 21 relationships+

The reference date can be a supplier article, integration announcement, customer record or letter of intent. It does not automatically identify a private purchase decision.

06 / THE LOOK-BACK

Find what changed around the relationship.

We checked five event families across the same companies: product updates, software work, operating teams, production preparation and geographic expansion. Each kept its source, date and limits.

21 relationshipsFive kinds of change
THE LOOK-BACKIN ACTION
BEFORE THE DEAL

We checked five event families across the same companies: product updates, software work, operating teams, production preparation and geographic expansion. Each kept its source, date and limits.

Dates and evidence+

Five event types and forty retained company-event records. Strong direct sources and weaker staffing or repository proxies remain distinguishable. A role start is not proof of factory output; an empty initial repository commit is not a functional software release.

07 / THE PATTERN TEST

Look for the work behind the announcement.

Software and AI work appeared in thirteen records; product changes in eleven. Six involved operating functions. These were openings for a more specific application question—not proof that a company wanted to buy a new hand.

40 company–event records17 of 21 companies represented
THE PATTERN TESTIN ACTION
ACROSS THE DEALS

Software and AI work appeared in thirteen records; product changes in eleven. Six involved operating functions. These were openings for a more specific application question—not proof that a company wanted to buy a new hand.

All five event types+

Counts describe retained evidence in this selected cohort. Forty records cover seventeen companies; a company can appear in several types. These are not purchase-conversion or proven causal statistics.

  • A new robot or product line goes out · 11 records (9 strong-source records)
  • An AI or robot-software stack goes in · 13 records (6 strong-source records)
  • An operating function is stood up · 6 records (0 strong-source records)
  • Production scales up · 4 records (2 strong-source records)
  • They enter a new country · 6 records (0 strong-source records)

08 / THE APPLICATION CONTEXT

Understand what the hardware is there to do.

The public accounts describe different jobs: collecting manipulation data, training dexterous policies and tending machines. We used those applications to qualify the opportunity. A proprietary hand programme needs a different approach from an open component platform.

A documented applicationA supplier route to qualify
THE APPLICATION CONTEXTIN ACTION
PUBLISHED TECHNICAL ACCOUNTS

The public accounts describe different jobs: collecting manipulation data, training dexterous policies and tending machines. We used those applications to qualify the opportunity. A proprietary hand programme needs a different approach from an open component platform.

The published accounts+

Published technical and supplier accounts, not private buyer-interview recordings. The application and the supplier relationship are kept separate where the source does not connect them.

Sanctuary AI · source account ↗

ABB Robotics · source account ↗

Automation Within Reach · source account ↗

PHASE02

Find the next programmes.

Look for product, software and operating changes—and check whether an external hand has a route in.

09 / FROM RESEARCH TO SOFTWARE

Give each change its own search.

A robot launch leaves different evidence from a manipulation repository or a service-team hire. We defined the sources and checks for each, then added the question that matters: can this programme use an external hand?

Five event familiesApplication + integration checks
FROM RESEARCH TO SOFTWAREIN ACTION
THE SEARCH BRIEF

A robot launch leaves different evidence from a manipulation repository or a service-team hire. We defined the sources and checks for each, then added the question that matters: can this programme use an external hand?

What the search checks+

The proposed watch workflow uses original product announcements, technical repositories, jobs and operating records. It checks sourcing openness and the application before treating an event as an account to approach.

10 / THE ACCOUNT OUTPUT

Find the next integration conversation.

The September shortlist contains twenty researched companies. Each connects an event, an application and the relevant team. We separated opportunities to pursue from accounts where the hardware sourcing route still needed validation.

20 researched companiesOpen route or validation needed
THE ACCOUNT OUTPUTIN ACTION
THE RESEARCHED ACCOUNT

The September shortlist contains twenty researched companies. Each connects an event, an application and the relevant team. We separated opportunities to pursue from accounts where the hardware sourcing route still needed validation.

The September shortlist+

Twenty companies were selected on 18 September 2026. The subsequent 22 September source pass refined interpretation; it did not re-verify all personnel or procurement status. Pursue/validate labels belong to that historical research snapshot.

PHASE03

Start with the integration question.

Use the specific platform and application to offer a useful first conversation.

11 / THE WAY INTO THE ACCOUNT

Make the first approach about their platform.

Universal Robots’ Gen 7 launch gave us a concrete starting point: an AI-ready tool flange and a new platform generation. The useful offer is an integration checklist—before promising compatibility or asking for a generic demo.

A named platform changeA practical integration checklist
THE WAY INTO THE ACCOUNTIN ACTION
THE COMPANY RESEARCH

Universal Robots’ Gen 7 launch gave us a concrete starting point: an AI-ready tool flange and a new platform generation. The useful offer is an integration checklist—before promising compatibility or asking for a generic demo.

The integration context+

The Gen 7 platform announcement and AI-ready tool flange are the account-specific starting point. A checklist is an offer to explore requirements, not a compatibility claim.

Universal Robots platform announcement ↗

Platform announcement ↗

12 / AN EXAMPLE FIRST EMAIL

Ask a question the engineering team can use.

This example turns that account research into a short first approach. It offers a checklist for a real integration question. It is a proposed message for review, not an email we claim was sent.

Platform + application contextA proposed first message
AN EXAMPLE FIRST EMAILIN ACTION
THE OUTREACH

This example turns that account research into a short first approach. It offers a checklist for a real integration question. It is a proposed message for review, not an email we claim was sent.

Read the complete example+

Proposed example created for this case study. Not sent or client-approved campaign copy. No reply, meeting, purchase or revenue outcome is claimed.

Gen 7 hand integration

hi {first_name},

your Gen 7 launch introduces an AI-ready tool flange and a new platform generation. i wondered how your application team is evaluating adaptive hands for tasks that involve varied objects.

a useful first check would cover mounting, power and control interfaces, then the objects, grip feedback and service needs of one reference application. that would help establish whether there is a fit before a hardware demo.

would a one-page integration checklist be useful for your application engineering team?

{sender}

Platform announcement ↗

THE POINT OF THE WORK

A better place to begin
the next sale.

A defined competitor set. Customer relationships with evidence.
Specific changes to look for. Something useful to say.

Find your starting point
Research → daily search → relevant conversations Built around the business you want to win.