Frequently asked questions
DataWise is a boutique data-science consultancy that builds production AI systems for Israeli mid-market manufacturers, importers, distributors, operations-heavy businesses, and security-sensitive organizations. DataWise prices in three ways: hourly advisory, fixed-scope-per-stage projects, and monthly support of delivered systems. Every engagement opens with a characterization stage that can honestly conclude “not yet.” You own the models, code, and pipelines at handover, with no platform meter and no lock-in. The questions below are the ones buyers actually ask before the first call.
How do you price your work?
DataWise prices three ways, matched to the need: hourly for advisory and specific questions, fixed-scope pricing per stage for projects, and a monthly retainer for ongoing support of delivered systems. Project pricing is quoted per stage, so you commit to characterization first, see its output, and only then decide about the feasibility stage. We do not publish a rate card because scope drives cost, but you will always know the full price of a stage before it starts, and each stage ends with a real decision point where you can stop.
What does a typical project cost?
We would be inventing a number if we gave you one before characterization, and inventing numbers is exactly what we tell clients to distrust. What we can say: the first stage is a bounded, fixed-price piece of work, deliberately sized so that finding out whether the project is worth doing is itself affordable. After it, you have a written scope and a firm quote for the next stage.
How long until we see results?
Honestly: it depends on your data readiness and system access, which is why we hedge. As a rough orientation, characterization typically runs a few weeks; a feasibility stage with real-data validation is typically measured in weeks to a few months, not days and not years. You see measurable evidence, results on your real data against the agreed success metrics, at the end of the feasibility stage, before committing to implementation.
Which systems do you integrate with?
The ones Israeli mid-market companies actually run. We have scoped and built against Priority, SAP, monday.com, and Hashavshevet (חשבשבת) environments, alongside data warehouses, spreadsheets that behave like databases, and everything in between. The integration plan is named explicitly in the characterization document: which systems, which direction data flows, and who signs off on access.
Can you work with sensitive data or inside a closed network?
Yes, closed-network and sensitive-data work is a deliberate DataWise specialty. For a security-sensitive organization DataWise architected and specified a fully isolated, zero-egress AI environment and built and demonstrated a self-contained AI assistant running entirely inside the closed network, with nothing leaving it: chat over organizational knowledge, retrieval over internal documents, document generation. If your constraint is "no internet, ever", that is a design requirement we know how to meet, not a dealbreaker.
Does our data get used to train anything outside our environment?
No. DataWise keeps your data in your environment, uses it only for your project, and never pools, resells, or uses it to improve anyone else's models. Where the architecture is fully self-hosted, the data physically cannot leave.
Who actually does the work?
DataWise is the founder, a senior data scientist and engineer, working directly. The person who sits in the characterization meetings is the person who writes the code and the person who hands the system over. There is no sales layer and no separate delivery team, so nothing is lost in a handover, because there is no handover.
What if the feasibility stage shows it does not work?
Then you stop, and that is the system working as designed. You keep the findings, the code produced to date, and a written account of why the numbers fell short and what would have to change. Success metrics are agreed before we build precisely so that this decision is made by numbers, not by sunk-cost momentum.
Do we own what you build?
Yes. At DataWise, the models, code, and pipelines are yours at handover. There is no platform meter, no per-prediction fee, and no dependency that stops the system if we part ways. Any competent engineer can maintain what we hand over, and the documentation is written with that person in mind.
What happens after delivery?
Your choice of two modes. Full handover: your team operates it, with documentation and a transition period. Or monthly support: we monitor performance, maintain the prediction-versus-actuals ledger, retrain models as reality drifts, and keep improving. Models degrade when the world changes, so if you choose full handover we will still tell you what to watch for.
Do you work in Hebrew and English?
Both, natively. Meetings, documentation, and deliverables are in whichever language your team works in, and engagements with Israeli companies serving foreign markets routinely run in both at once. AI systems we build can serve Hebrew-speaking users, including Hebrew voice and text interfaces.
How ready do our data need to be?
Less ready than you fear, but the honest answer comes from looking, which is what characterization is for. You do not need a data team or a warehouse project before talking to us. You do need real historical data that reflects the business. If the data are not ready, we will say so and tell you what to fix first; that conclusion costs a characterization, not a failed project.
Do you build chatbots?
Not website chatbots; off-the-shelf tools do that well and cheaply, and we will point you to that path. We do build conversational systems where they carry real operational weight: organizational knowledge agents over internal content, and voice or WhatsApp agents wired into a company's actual content and workflows.
What size and kind of company do you work with?
Mostly Israeli mid-market companies in manufacturing, distribution, import, and operations-heavy services, plus security-sensitive organizations with closed-network requirements. If you need an organization-wide transformation program or a large staff-augmentation bench, we are the wrong size for you on purpose, and we will say so in the first call.