Why Your Marketing ROI Numbers Don’t Match Across Platforms

Your Meta dashboard says 120 conversions. Your CRM says 80. Your billing system says 42. None of them are broken — they're just measuring different things.

The culprit is usually the attribution window itself, not a broken pixel or tracking misconfiguration, as one attribution analysis explains. Meta credits conversions within a 7-day click window, while HubSpot's lookback stretches to 30 days, distributing credit across a broader set of touchpoints. The same campaign can look like a star in one report and a dud in another.

One documented B2B case captures the chaos perfectly: Google Analytics reported 500 conversions, the CRM showed 380 leads, billing counted 42 new customers — yet ad platforms collectively claimed credit for over 700. Every board meeting became a debate about which numbers were "real," and trust in the data eroded to zero.

Why the gaps happen:

• Different default windows — Meta defaults to 7-day click + 1-day view; HubSpot recommends 30-day cross-channel windows, so the same conversion gets counted differently (per HubSpot's guidance).
• Self-reporting bias — every platform "claims as much credit as possible for itself," meaning you're relying on numbers from channels with an obvious incentive to over-report (according to AppsFlyer).
• Modeled vs. observed data — for iOS-heavy audiences, 40–60% of conversions in a 7-day click window are statistical estimates, not pixel-confirmed counts (research shows).
• No shared definition of revenue — marketing measures pipeline, finance measures cash collected, sales measures bookings. Three valid metrics, three different numbers.

The consequences go beyond reporting headaches. Only 29% of marketers are extremely confident in their attribution, and misaligned windows distort budget decisions — longer windows inflate credited conversions while shorter windows make lower-funnel channels look stronger. When the CFO and CMO bring conflicting reports to the same meeting, budget allocation stalls.

For service businesses running reactivation campaigns alongside acquisition ads, this matters practically. A win-back call that books an appointment weeks after a seasonal reminder may get credited to whichever platform holds the longest window — even when the customer came back because someone actually picked up the phone. CallMyCustomers sees this with client lists pulled straight from CRMs and point-of-sale systems: the booked job is the number that counts, not what any ad platform claims.

The fix starts with documenting which window each platform uses, aligning window length to your actual sales cycle, and treating cross-platform numbers as directional rather than gospel.

How to Choose the Right Attribution Window for Your Service Business Sales Cycle

Choosing the right attribution window isn’t about default settings—it’s about matching your actual customer behavior. For service businesses like HVAC, dental clinics, or auto repair shops, the time between a reactivation touchpoint and a booked appointment often follows predictable patterns rooted in seasonal needs, service cycles, or customer inertia. Relying on platform defaults like Meta’s 7-day click window or HubSpot’s 30-day lookback can distort your view of what’s truly driving repeat revenue, especially when campaigns aim to re-engage customers who haven’t booked in 6, 12, or even 18 months.

Start by analyzing your CRM deal velocity data—the average time from first outreach to booked job for reactivated customers. Research shows that teams who align attribution windows with real sales cycle data avoid misconfigurations that distort revenue trends and lead-quality interpretation. For home service win-back campaigns, this might reveal a 30- to 60-day window as most accurate, while clinic membership renewals could show stronger attribution at 90 days due to longer decision cycles. Testing multiple window lengths—such as comparing 7-day, 30-day, and 60-day settings—lets you see which best reflects actual booking patterns and prevents over- or under-crediting specific channels.

• Track how long it takes customers to book after receiving a seasonal reminder or quote follow-up—this defines your baseline velocity.
• Compare attributed conversions across window lengths to identify where credit is being misallocated (e.g., short windows ignoring nurture influence).
• Document your chosen window in a reporting playbook so team members and platforms apply it consistently.
• Pair attribution data with incrementality testing to confirm whether credited touchpoints actually drove the booking.

For CallMyCustomers, this means ensuring that when measuring the impact of a win-back or membership renewal campaign, the attribution window reflects the true time it takes a past customer to respond—not just what a platform assumes. A window too short might miss the influence of a well-timed text reminder that led to a call two weeks later; too long could credit a touchpoint for a booking that was already inevitable. By grounding your window in real CRM data—like the average days between outreach and appointment for reactivated HVAC or dental patients—you gain clarity on which campaigns genuinely move the needle. This alignment transforms attribution from a reporting exercise into a tool for smarter budget decisions, especially when proving the ROI of reactivation as a second revenue engine alongside new lead acquisition.

Practical Steps to Test, Validate, and Improve Your Attribution Setup

Testing your attribution setup requires a structured approach to ensure accuracy and alignment with real customer behavior. Start by establishing a baseline using your current window settings, then run parallel tests with alternative durations—such as comparing a 7-day click window against a 30-day lookback—to measure differences in attributed conversions, ROAS, and channel performance. This method helps identify whether your current configuration over- or under-credits specific touchpoints, especially important for reactivation campaigns where timing between outreach and booking can vary. For CallMyCustomers’ client approval workflow, documenting these tests alongside script and offer sign-offs ensures transparency and data-driven decision-making throughout the campaign lifecycle.

Pairing attribution data with incrementality testing is essential to validate whether credited conversions were genuinely influenced by your outreach. As research shows, attribution tells you who was credited, but incrementality reveals whether the credit was earned—preventing costly misallocations based on correlated but not causal activity. Running holdout tests, where a control group receives a neutral message instead of your reactivation offer, allows you to measure true lift. For example, if 7-day attributed conversions show 100 bookings but holdout testing reveals only 40 incremental conversions, your window may be overclaiming by 2.5×, signaling a need to shorten the window or refine targeting.

Improving data hygiene strengthens the foundation of your attribution model. Inconsistent UTM tracking, fragmented CRM data, or poor list segmentation can limit attribution coverage—research indicates that fixing these issues can increase coverage from ~60% to 95%. For CallMyCustomers’ process, this means ensuring lists imported from spreadsheets, CRMs, or POS systems are properly cleaned, tagged by recency (e.g., 30 days, 6 months, 12+ months), and tracked through every stage—from list review and message approval to outreach, booking, and post-service follow-up. Clean, integrated data ensures that when a customer books after a text or call, the conversion is accurately attributed to the correct touchpoint within your chosen window.

Finally, align your attribution window length with actual sales cycle data from your CRM. Rather than relying on platform defaults, analyze historical deal velocity to determine the typical time between marketing touchpoints and booked appointments. For US service businesses in home services, dental clinics, or automotive repair, this often falls within a 7–30 day range for reactivation efforts, though seasonal or membership-based campaigns may require longer windows. Testing multiple lengths and documenting the chosen approach in a reporting playbook creates consistency across campaigns and client approvals, turning attribution from a reporting exercise into a reliable tool for estimating revenue impact and optimizing reactivation ROI.